Analyseit
http://analyseit.com/blog/
Analyseit statistical analysis and visualization software blog. Provides the latest news, releases, upcoming new features and informative articles so you get the most out of Analyseit and Microsoft Excel.
engb
© 2017 Analyseit Software, Ltd.
Subscribe with My Yahoo!Subscribe with BloglinesSubscribe with Google
https://analyseit.com/blog/2017/5/parameterestimation
Parameter estimation
http://feeds.analyseit.com/~r/Analyseit/~3/qQbnV_wgWJI/parameterestimation
<div class="article">
<div class="content">
<p>Often we collect a sample of data not to make statements about that particular sample but to generalize our statements to say something about the population. Estimation is the process of making inferences about an unknown population parameter from a random sample drawn from the population of interest. An estimator is a method for arriving at an estimate of the value of an unknown parameter. Often there are many competing estimators for the population parameter that differ based on the underlying statistical theory. </p>
<h2>Point estimate</h2>
<p>A point estimate is the best estimate, in some sense, of the population parameter. The most wellknown estimator is the sample mean which produces an estimate of the population mean.</p>
<p>It should be obvious that any point estimate is not absolutely accurate. It is an estimate based on only a single random sample. If repeated random samples were taken from the population the point estimate would be expected to vary from sample to sample. This leads to the definition of an interval estimator which provides a range of values defined by the limits [L, U].</p>
<h2>Confidence interval</h2>
<p>A confidence interval defines limits [L, U] constructed on the basis that a specified proportion of the confidence intervals include the true parameter in repeated sampling. How frequently the confidence interval contains the parameter is determined by the confidence level. 95% is commonly used and means that in repeated sampling 95% of the confidence intervals include the parameter. 99% is sometimes used when more confidence is needed and means that in repeated sampling 99% of the intervals include the parameter. It is unusual to use a confidence level of less than 90% as too many intervals would fail to include the parameter. </p>
<p>Again a confidence interval is formed using an interval estimator based on statistical theory and assumptions about the underlying population. The tbased confidence interval estimator for a population mean is an example taught in most introductory textbooks.</p>
<p>Many people misunderstand confidence intervals. A confidence interval is a frequentist concept, that is, the probability is defined in a series of repetitions of an experiment. <b>A confidence interval does not predict with a given probability that the parameter lies within the interval.</b> The problem arises because the word confidence is misinterpreted as implying probability. After drawing a sample and forming a confidence interval, it either does contain the population parameter or it doesn’t. This is the same as how the probability of obtaining a head or tails in the toss of fair coin is 50%, but after a toss has happened it is either a head or tail (the predicted event happened or it didn’t). Similarly, the 95% probability associated with a 95% confidence interval only applies to the method used to construct the interval, not to the individual realized intervals. </p>
<p>As an example, let’s say we draw a single random sample and construct a 95% confidence interval for the mean that results in an interval [15, 20]. We could say “We are 95% <i>confident</i> the population mean is between 15 and 20”. It would be incorrect to say “There is a 95% <i>probability</i> the population mean is between 15 and 20”. The switch here is a subtle change of the word confident to the word probability. The word confident means that the estimation method works 95% of the time, the other 5% of the time it is wrong. </p>
<h2>Illustration</h2>
<p>We can illustrate these concepts with a MonteCarlo simulation. </p>
<p>Let’s assume the population has a mean μ=0 and standard deviation σ=1, and we draw 100 random samples from the population each with a size n=40, then compute the 95% mean confidence interval for each sample.</p>
<p><img src="https://analyseit.com/assets/article/54611645/MeanConfidenceIntervalSimulation.png" alt="Mean confidence interval simulation: Analyseit blog"></p>
<p>The plot shows the limits of the confidence interval as an error bar, with the mean as a dot, for each of 100 samples. The horizontal red line indicates μ=0, the true population mean. Highlighted in red are repetitions where the mean confidence interval does not include μ=0. You can see that 5/100=5% are colored red and 95/100=95% are colored black, and therefore interval included the population mean 95% of the time. Proof that our estimation method works 95% of the time!</p>
<h2>Reporting results</h2>
<p>A point estimate of the parameter from a single sample is not a great way to present your findings. It gives no idea about uncertainty. A point estimate with a confidence interval provides more information, as the point estimate is the most likely value given the sample observed and the confidence interval expresses the uncertainty in this estimate. If the width of the confidence interval is large you should have less confidence in the estimate, and if the interval is narrow you can have more confidence. When the interval is wide you could repeat the experiment with a larger sample size to get a narrower interval. Repeating an experiment is always a good way to confirm that your results are real and is key to making good scientific judgments. </p>
<p><b><br>
</b></p>
<p><b>Related links:</b></p>
<p>A great interactive visualization of the confidence interval: <a href="http://rpsychologist.com/d3/CI/" target="_blank">http://rpsychologist.com/d3/CI/</a></p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/qQbnV_wgWJI" height="1" width="1" alt=""/>
Using Analyseit
Statistics
Thu, 18 May 2017 15:40:39 GMT
https://analyseit.com/blog/2017/5/parameterestimation

https://analyseit.com/blog/2017/2/analyseit480indepththerevampeddocumentation
Analyseit 4.80 indepth: The revamped documentation
http://feeds.analyseit.com/~r/Analyseit/~3/IZhPVEeh_UA/analyseit480indepththerevampeddocumentation
<div class="article">
<div class="content">
<p>As we mentioned last week in the <a href="https://analyseit.com/blog/2017/2/analyseit480releasedanovaancovaandrevampeddocumentation" target="">post announcing Analyseit 4.80</a>, in this release we took the opportunity to revamp the documentation.</p>
<p>The revamp involved rewriting many topics to make the content clearer, adding new taskoriented topics, including refresher topics on common statistical concepts, and improving the indexing and links between topics so you can more easily navigate the help system. </p>
<p>The new taskoriented topics give you stepbystep instructions on completing common tasks. For example you will now find topics on how to <a href="https://analyseit.com/docs/userguide/processcontrol/creatingshewhartcontrolchartsubgroup" target="">create an XbarR Shewhart control chart</a>, <a href="https://analyseit.com/docs/userguide/fitmodel/linear/performinganova" target="">how to perform ANOVA</a>, <a href="https://analyseit.com/docs/userguide/fitmodel/linear/fittingadvancedmodels" target="">how to fit advanced models</a>, and even simple tasks like <a href="https://analyseit.com/docs/userguide/distribution/continuous/creatingunivariatedescriptives" target="">calculating descriptive statistics</a>. We have also fully documented the supported dataset layouts for each type of analysis so you can see how to arrange your data for Analyseit. The links in each topic help you more easily find related topics, for example links to topics on how to interpret the statistics, links to explain the pros and cons of the available statistical tests, links to topics for common tasks, and a link showing you how to arrange the dataset.</p>
<p>Of all the requests, the <b>most customerrequested improvement</b> is the new <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.pdf" target="">printable PDF user guide</a>. Previously we offered EPUB and Kindle reader editions of the help, but not PDF. To be honest, producing a PDF user guide from the <a href="https://en.wikipedia.org/wiki/Darwin_Information_Typing_Architecture" target="">DITA</a> tools we use to write the help was a real technical challenge. The PDF produced just wasn’t good enough for us, and certainly not for our customers – the formatting and layout were poor, indexing was nonexistent, and there were so many other niggles. So we took the time to make the userguide both look good and be usable. Take a look and let us know what you think!</p>
<p>With the addition of the new PDF user guide, you can now find documentation in a number of formats suitable in all situations:</p>
<ul>
<li><b>Online:</b> <a href="https://analyseit.com/support/" target="">Browse and search the documentation</a> online.<br>
</li>
<li><b>In application:</b> Access the help directly from within Analyseit, even when you’re offline. To see the documentation, click the Analyseit icon on the Analyseit ribbon tab, then click Help. The Microsoft HTML Help will be shown, complete with full index and search capabilities. You can also see the contextsensitive help popup for any option or ribbon command in Analyseit simply by hovering the mouse over it.<br>
</li>
<li><b>For printing or reading on your computer:</b> To get a fully formatted book with table of contents and index that you can read or print, <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.pdf" target="">download the PDF</a>.<br>
</li>
<li><b>For reading offline on an iPad or Kindle device:</b> For best reading experience on an <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.epub" target="">iPad download the EPUB</a>, or on a <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.mobi" target="">Kindle device download the MOBI</a>. The eBook versions are more flexible than the PDF as the content is reformatted and reflows depending on the screen size of your device and your preferred font size.<br>
</li>
</ul>
<p>Let us know your thoughts and how we can further improve the documentation.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/IZhPVEeh_UA" height="1" width="1" alt=""/>
Releases
Using Analyseit
Tue, 28 Feb 2017 15:02:26 GMT
https://analyseit.com/blog/2017/2/analyseit480indepththerevampeddocumentation

https://analyseit.com/blog/2017/2/analyseit480releasedanovaancovaandrevampeddocumentation
Analyseit 4.80 released: ANOVA, ANCOVA and revamped documentation
http://feeds.analyseit.com/~r/Analyseit/~3/pKiRtKGousA/analyseit480releasedanovaancovaandrevampeddocumentation
<div class="article">
<div class="content">
<p><img src="https://analyseit.com/assets/article/54611643/ANOVAANCOVAmenu.jpg" style="float:right; paddingleft: 16px;">Last week we released version 4.80 of Analyseit.</p>
<p>The new release includes multiway <a href="https://analyseit.com/docs/userguide/fitmodel/linear/performinganova" target="">ANOVA</a>, <a href="https://analyseit.com/docs/userguide/fitmodel/linear/performingancova" target="">ANCOVA</a>, and <a href="https://analyseit.com/docs/userguide/multivariate/itemreliability" target="">Cronbach’s Alpha</a> in the Standard edition, and since every licence includes the Standard edition, these features are available to all users. We also took the opportunity to revamp the <a href="https://analyseit.com/support/" target="">help for Analyseit</a> and develop a <a href="https://analyseit.com/docs/Analyseit_CompleteGuide.pdf" target="">PDF userguide</a>. We’ll go into more details on the improvements in the next few weeks.</p>
<p>If you have <a href="http://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="http://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="http://analyseit.com/store/maintenance/" target="_blank"> renew maintenance</a>.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/pKiRtKGousA" height="1" width="1" alt=""/>
Using Analyseit
Releases
Tue, 07 Feb 2017 11:47:26 GMT
https://analyseit.com/blog/2017/2/analyseit480releasedanovaancovaandrevampeddocumentation

https://analyseit.com/blog/2016/1/analyseit460released3nestedfactorprecision
Analyseit 4.60 released: 3 nested factor precision
http://feeds.analyseit.com/~r/Analyseit/~3/DR2wbwy2WRA/analyseit460released3nestedfactorprecision
<div class="article">
<div class="content">
<p>Today we released version 4.60 of the Analyseit Method Validation edition.</p>
<p>The new release now includes 3 nestedfactor precision analysis, which extends Analyseit’s support for CLSI EP05A3 multilaboratory precision studies.</p><p><img src="https://analyseit.com/assets/article/54611642/3wayPrecision.png" style="cursor: default;"></p>
<p>If you have <a href="http://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="http://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your license has expired you can renew it to get this update and forthcoming updates, see <a href="http://analyseit.com/store/maintenance/" target="_blank"> renew maintenance</a>.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/DR2wbwy2WRA" height="1" width="1" alt=""/>
Method validation
Releases
Thu, 21 Jan 2016 12:20:27 GMT
https://analyseit.com/blog/2016/1/analyseit460released3nestedfactorprecision

https://analyseit.com/blog/2015/10/announcingtheanalyseitqualitycontrolimprovementedition
Announcing the Analyseit Quality Control & Improvement Edition
http://feeds.analyseit.com/~r/Analyseit/~3/vpRnjXvS8f0/announcingtheanalyseitqualitycontrolimprovementedition
<div class="article">
<div class="content">
<h3><b>Update 15Dec2015</b></h3>
<p>We have just released an update to the prerelease. The software is now available for anyone to test, and it now includes tutorials you can follow to quickly get started using Pareto analysis, Control charts, and Capability analysis. It does not require an active internet connection as we no longer need to monitor the software for error conditions.</p>
<p>You can download the latest release and try the software from:</p>
<p><a href="http://analyseit.com/download/Analyseit/4.51/Analyseit_4_51.EXE">http://analyseit.com/download/Analyseit/4.51/Analyseit_4_51.EXE</a></p>
<p>To activate the use the product key:</p>
<pre>PTESTD4NF4LXJSYTT
</pre>
<h3><b>Original article</b></h3>
<p>We are delighted to announce the addition of the <b>Analyseit Quality Control and Improvement Edition</b> to the range of Analyseit products.</p>
<p>The new edition includes the most impressive statistical process control (SPC) charts available in any Excel statistical software package, including Shewhart, LeveyJennings, CUSUM, and EWMA charts. Process capability statistics and plots help you ensure a process is able to meet specifications. And Pareto plots help you identify the qualityrelated problems that need the most attention and let you monitor efforts to reduce them. </p>
<p><i>Screenshot: XbarR chart of a quality characteristic by phase with stratification.</i></p>
<p><img src="https://analyseit.com/assets/article/54611641/controlchart2.png" style="cursor: nwresize;"></p>
<p><i>Screenshot: Process capability plots by phase.</i></p>
<p><img src="https://analyseit.com/assets/article/54611641/capability.png" style="cursor: default;"></p>
<p><i>Screenshot: Pareto plot of failures with stratification and sorted by key plot.</i></p>
<p><img src="https://analyseit.com/assets/article/54611641/pareto.png" style="cursor: default;"></p>
<h2>Download the prerelease now</h2>
<p>The new release is currently available for prerelease beta testing. Anyone can download and test the new release, though you will need an active internet connection to use it as we monitor usage and reliability of the prerelease.</p>
<p>You can download the prerelease at:</p>
<p><strike>http://analyseit.com/download/Analyseit/4.50/Analyseit_4_50.EXE</strike></p>
<p>To activate the prerelease, use the product key:</p>
<pre>PTESTD4NF4LXJSYTT
</pre>
<br>
<p>We are now working to update the documentation to explain the new features of the software, but in the meantime these examples should help you get started</p>
<p><a href="https://analyseit.com/assets/article/54611641/Pareto.xlsx" class="excelfile">Pareto.xlsx: Pareto plots of failures stratified by date and before/after maintenance</a></p>
<p><a href="https://analyseit.com/assets/article/54611641/BurstingStrength.xlsx" class="excelfile">BurstingStrength.xlsx: Xbar R control charts and capability analysis of bursting strength of glass containers</a></p>
<p><a href="https://analyseit.com/assets/article/54611641/Cans.xlsx" class="excelfile">Cans.xlsx: Shewhart p and np control plots analysis of can seal attribute data</a></p>
<p>We will provide more information about the product, availability and pricing shortly, including pricing for existing users that want to upgrade to include the new quality control features. </p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/vpRnjXvS8f0" height="1" width="1" alt=""/>
Releases
Business
In development
Tue, 20 Oct 2015 13:05:31 GMT
https://analyseit.com/blog/2015/10/announcingtheanalyseitqualitycontrolimprovementedition

https://analyseit.com/blog/2015/9/analyseit420releasedsupportforexcel2016
Analyseit 4.20 released: Support for Excel 2016
http://feeds.analyseit.com/~r/Analyseit/~3/XYt7nn0cXUM/analyseit420releasedsupportforexcel2016
<div class="article">
<div class="content">
<p><img src="https://analyseit.com/assets/article/54611640/Office2016sm.png" style="float:right">
Microsoft officially released <a href="https://products.office.com/" target="_blank">Office 2016</a> a couple of days ago, and Analyseit version 4.20 now adds support for Excel 2016. </p>
<p>Over the next few weeks we will tweak the Analyseit user interface so it matches the new Office 2016 user interface styles. Line styles on the plots in Excel 2016 now also appear a lot thicker, due to antialiasing (smoothing), so we will decide whether to address that in a future update – let us know what you think.</p>
<p>If you have <a href="http://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="http://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired you can renew it to get this update and forthcoming updates, see <a href="http://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/XYt7nn0cXUM" height="1" width="1" alt=""/>
Releases
Excel
Thu, 24 Sep 2015 11:15:17 GMT
https://analyseit.com/blog/2015/9/analyseit420releasedsupportforexcel2016

https://analyseit.com/blog/2015/8/thenumericalaccuracyofanalyseitagainsttheniststrd
The numerical accuracy of Analyseit against the NIST StRD
http://feeds.analyseit.com/~r/Analyseit/~3/LTKLSfy5q3U/thenumericalaccuracyofanalyseitagainsttheniststrd
<div class="article">
<div class="content">
<p>A critical feature of any analytical and statistical software is accuracy. You are making decisions based on the statistics obtained and you need to know you can rely on them.</p>
<p>We have documented our <a href="https://analyseit.com/blog/2014/1/oursoftwaredevelopmentandvalidationprocess">software development and validation process</a> previously, but another good benchmark to test statistical software against is the NIST StRD. The Statistical Engineering and Mathematical and Computational Sciences Divisions of NIST’s Information Technology Laboratory developed datasets with certified values for a variety of statistical methods against which statistical software packages can be benchmarked. <span style="fontfamily: Arial, Helvetica, Verdana, Tahoma, sansserif; fontsize: 15px; lineheight: 1.45em; backgroundcolor: initial;">The certified values are computed using ultrahigh precision floating point arithmetic and are accurate to 15 significant digits.</span></p>
<p>For more information about the NIST StRD see:</p>
<img src="http://www.itl.nist.gov/div898/strd/general/gifs/strd80.gif">
<p><a href="http://www.itl.nist.gov/div898/strd/" target="_blank">http://www.itl.nist.gov/div898/strd/</a></p>
<p><span style="fontfamily: Arial, Helvetica, Verdana, Tahoma, sansserif; fontsize: 15px; lineheight: 1.45em; backgroundcolor: initial;">We tested version 4.00 of Analyseit against the NIST StRD on an Intel Xeon dual processor PC.</span><br>
</p>
<p>No statistical package achieves perfect accuracy for all the tests and no one package performs best for every test. Most statistical packages use IEEE754 double precision (64bit) floating point arithmetic and due to finite precision, roundoff, and truncation errors in numerical operations, are unable to obtain the exact certified value.</p>
<p>In the tests:</p>
<ul>
<li><b>Analyseit performed consistently amongst the best on all tests.</b></li>
<li><b>Analyseit performed better than some of the most popular wellknown statistical packages</b>.</li>
</ul>
<p>For more information on the performance of Analyseit against the NIST StRD, and to download the workbooks containing the analyses, see:</p>
<p><a href="https://analyseit.com/support/NISTStRD">http://analyseit.com/support/NISTStRD</a></p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
<a id="arss_hlComments" href="/blog/2015/8/thenumericalaccuracyofanalyseitagainsttheniststrd#comments">1 comment</a>
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/LTKLSfy5q3U" height="1" width="1" alt=""/>
Statistics
Business
Tue, 11 Aug 2015 11:36:43 GMT
https://analyseit.com/blog/2015/8/thenumericalaccuracyofanalyseitagainsttheniststrd

https://analyseit.com/blog/2015/6/analyseit40releasedsupportforclsiguidelinesandmeasurementsystemsanalysis
Analyseit 4.0 released: Support for CLSI guidelines, and Measurement Systems Analysis
http://feeds.analyseit.com/~r/Analyseit/~3/FJEyJzAnqhQ/analyseit40releasedsupportforclsiguidelinesandmeasurementsystemsanalysis
<div class="article">
<div class="content">
<p>Today we released version 4.0 of the Analyseit Method Validation edition. This is a major new release with many new features and improvements.</p>
<h2>Support for 10 of the latest CLSI EP guidelines</h2>
<p>The latest release of the Analyseit Method Validation edition now supports 10 of the latest CLSI evaluation protocol (EP) guidelines. <a href="http://clsi.org" target="_blank">CLSI</a> guidelines are worldrenowned and are recognized by the College of American Pathologists (CAP), The Joint Commission, and the US Food and Drug Administration (FDA).</p>
<p>Analyseit has been a driving force in the adoption of statistics in method validation for over 15 years, has influenced many recommendations, and is the only software available with such extensive coverage for the latest CLSI guideles.</p>
<p>CLSI guidelines supported in version 4.0 include:</p>
<table class="table small">
<tbody>
<tr>
<td class=""><a href="http://shop.clsi.org/methodevaluationdocuments/EP05.html" target="_blank">EP05A3</a><br>
Evaluation of Precision of Quantitative Measurement Procedures </td>
</tr>
<tr>
<td class="redactorcurrenttd"><a href="http://shop.clsi.org/methodevaluationdocuments/EP06.html" target="_blank">EP06A</a><br>
Evaluation of the Linearity of Quantitative Measurement Procedures </td>
</tr>
<tr>
<td class=""><a href="http://shop.clsi.org/methodevaluationdocuments/EP09.html" target="_blank">EP09A3</a><br>
Measurement Procedure Comparison and Bias Estimation Using Patient Samples </td>
</tr>
<tr>
<td><a href="http://shop.clsi.org/methodevaluationdocuments/EP10.html" target="_blank">EP10A3AMD</a><br>
Preliminary Evaluation of Quantitative Clinical Laboratory Measurement Procedures </td>
</tr>
<tr>
<td><a href="http://shop.clsi.org/methodevaluationdocuments/EP12.html" target="_blank">EP12A2</a><br>
User Protocol for Evaluation of Qualitative Test Performance </td>
</tr>
<tr>
<td><a href="http://shop.clsi.org/methodevaluationdocuments/EP15.html" target="_blank">EP15A3</a><br>
User Verification of Precision and Estimation of Bias </td>
</tr>
<tr>
<td class=""><a href="http://shop.clsi.org/methodevaluationdocuments/EP17.html" target="_blank">EP17A2</a><br>
Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures </td>
</tr>
<tr>
<td><a href="http://shop.clsi.org/methodevaluationdocuments/EP21.html" target="_blank">EP21A</a><br>
Estimation of Total Analytical Error for Clinical Laboratory Methods </td>
</tr>
<tr>
<td class=""><a href="http://shop.clsi.org/methodevaluationdocuments/EP24.html" target="_blank">EP24A2 (Replaces GP10A)</a><br>
Assessment of the Diagnostic Accuracy of Laboratory Tests Using Receiver Operating Characteristic Curves </td>
</tr>
<tr>
<td class=""><a href="http://shop.clsi.org/methodevaluationdocuments/EP28.html" target="_blank">EP28A3C (Formerly C28A3C)</a><br>
Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory </td>
</tr>
</tbody>
</table>
<h2>Measurement Systems Analysis (MSA)</h2>
<p>Measurement Systems Analysis (MSA) is a new feature in version 4.0. MSA unifies precision and linearity, which were available in earlier versions of Analyseit, but also includes trueness (bias) and detection capability so you can establish the limit of blank (LoB) and limit of detection (LoD). The unification of these analyses in MSA lets you dig deep to examine and understand the performance characteristics of a measurement procedure.</p>
<p><img src="https://analyseit.com/assets/article/54611638/Precision profile.PNG" style="cursor: default;"></p>
<p><img src="https://analyseit.com/assets/article/54611638/Trueness.PNG" style="cursor: default;"></p>
<p><img src="https://analyseit.com/assets/article/54611638/Detection limit.PNG" style="cursor: default;"></p>
<h2>Partition Method Comparison and Reference Interval analyses</h2>
<p>A major improvement in version 4.0 is the ability to split a method comparison analysis into specific measuring intervals and apply a different analysis. For example, you might split the analysis into two measuring intervals: samples <1ug/L, and those above. For <1ug/L the relationship between measurements might be different from those >1 ug/L, so you can apply different fits and goals to each measuring interval to see how the methods compare.</p>
<p>Similarly, for reference intervals you can apply partitioning factors to generate reference intervals for different subgroups. For example, you might apply partitioning factors sex and age group so you can determine reference intervals for age groups 020, 2035, 3550, and 50+ for male / female subjects. For each subgroup you can change how the reference interval is calculated, for example using the robust reference interval calculation for a subgroup with very few subjects.</p>
<p><img src="https://analyseit.com/assets/article/54611638/Method comparision.PNG" style="cursor: default;"></p>
<h2>Availability</h2>
<p>If you already have a licence for Analyseit, and you have active maintenance (access to software updates), you can download the latest version now. If maintenance on your licence has lapsed, now is a good time to renew maintenance to get this update and future updates before the price increase (see below). </p>
<p>To download the update, check whether you have maintenance, or to renew maintenance, see:</p>
<p><a href="http://analyseit.com/support/download" target="_blank">http://analyseit.com/support/download</a></p><p>If you are new to Analyseit you can download a free 30day trial at:</p><p><span style="fontfamily: Arial, Helvetica, Verdana, Tahoma, sansserif; fontsize: 15px; lineheight: 1.45em; backgroundcolor: initial;"><a href="http://analyseit.com/products/methodvalidation/try">http://analyseit.com/products/methodvalidation/try</a></span></p>
<h2>Pricing and licensing changes from 7th July 2015</h2>
<p>The cost of a 1user perpetual licence for the Analyseit Method Validation will increase from US$699 to US$749, and the cost of a 1 floatinguser perpetual licence will crease from US$1239 to US$1329 from the 7th July 2015. Since maintenance is based on a percentage of the licence price, the 1year maintenance renewal price for a 1user perpetual licence will increase slightly from US$69.90 to US$74.90.</p>
<p>We will also be trialling a new option to purchase a 1year annual subscription for the Analyseit Method Validation edition from the 7th July 2015. The cost will be US$ 449 for 1year and includes all software updates released within the year. At the end of the year you can renew your licence to continue using the software, or let the licence lapse if you no longer need it.</p>
<h2>Change summary</h2>
<p>Finally, below is a list of the major changes and improvements in version 4.0.</p>
<p><b>Reference interval</b></p>
<ul>
<li>Partition factors for separate reference interval analysis of subgroups.</li>
<li>Additional transformation functions: Manly exponential, 2stage Exponential/Modulus transformation.</li>
<li>Normal MVUE quantile estimator for reference limits.</li>
</ul>
<p><b>Diagnostic performance</b></p>
<ul>
<li>BiHistogram and dotpot of the distribution of positive/negative cases.</li>
</ul>
<p><b>Method comparison</b></p>
<ul>
<li>Adjust measuring interval to a linear range and extrapolate fits beyond the data to the measuring interval.</li>
<li>Partition into measuring intervals, with different relationships (e.g. constant / relative differences or variances) for each measuring interval.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>Improved report and plots.</li>
<li>Integrated into MSA to allow unified approach to evaluation of other components of error.</li>
</ul>
<p><b>Precision</b></p>
<ul>
<li>Improved report and plots.</li>
<li>Precision profile variance function fits with a choice of 7 fits, including Sadler 3parameter functions.</li>
<li>Limit of quantitation via prediction from CV% on precision profile plot.</li>
</ul>
<p><b>Trueness <span class="label labelimportant">NEW</span></b></p>
<ul>
<li>Estimate bias with reference materials or PT/EQA material. </li>
</ul>
<p><b>Detection capability <span class="label labelimportant">NEW</span></b></p>
<ul>
<li><span style="backgroundcolor: initial;">Estimate the decision level critical value (LoB) using parametric and nonparametric estimator on blank material or via the precision profile variance function.</span><br></li>
<li>Estimate the detection limit (LoD) using nonblank material or via the precision profile variance function. </li>
</ul>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/FJEyJzAnqhQ" height="1" width="1" alt=""/>
Method validation
Releases
Tue, 23 Jun 2015 14:08:02 GMT
https://analyseit.com/blog/2015/6/analyseit40releasedsupportforclsiguidelinesandmeasurementsystemsanalysis

https://analyseit.com/blog/2014/10/asombrenoteprofessorrickjones
A sombre note: Professor Rick Jones
http://feeds.analyseit.com/~r/Analyseit/~3/c59aNMMnJhc/asombrenoteprofessorrickjones
<div class="article">
<div class="content">
<p>The recent of passing of Professor Rick Jones (see <a href="http://www.theguardian.com/education/2014/sep/09/rickjonesobituary" target="_blank">Rick Jones obituary</a>) caused me to reflect on the past.</p>
<p>I was very fortunate to earn a work placement with Dr Rick Jones at The University of Leeds in the summer of 1990. Rick was enthusiastic about the role of IT in medicine, and after securing funding for a fulltime position he employed me as a computer programmer. Early projects included software for automating the monitoring of various blood marker tests and software to diagnose Down’s syndrome. At the time many hospitals had inhouse solutions for diagnosing Down’s syndrome, and although the project took many years and the help of many other people to complete, it eventually gained widespread adoption.</p>
<p>Around 1992, Rick came up with the idea of a statistics package that integrated into Microsoft Excel. Armed with a ring bound folder containing the Excel SDK and a pile of medical statistics books, I set about the task of writing the software in C++. It wasn’t long before the first version of Astute was ready and commercially released.</p>
<p>Never short of ideas for new projects, Rick started leaving journal articles in my intray covering subjects such as Deming and PassingBablok regression, GalenGambino sensitivity/specificity and ROC analysis. Rick had widespread experience and knowledge of the statistics needed in the clinical laboratory, and was keen to make the subject less daunting and more accessible to clinicians. The plan was to extend Astute to include these new statistical techniques. It would be an entirely new type of statistics package, so new in fact that when we released the product we didn’t have a good name for it and it was simply called “Astute  Module 1”. </p>
<p><img src="https://analyseit.com/assets/article/54611637/AstuteCover.JPG"> <img src="https://analyseit.com/assets/article/54611637/AstuteMod1Contents.JPG"></p>
<p>Rick later coauthored the book <a href="http://www.amazon.co.uk/InvestigationStatisticsLaboratoryManagementTechnology/dp/0902429213" target="_blank">Clinical Investigation and Statistics in Laboratory Medicine</a> and it helped many clinicians better understand and appreciate the use of statistics in the laboratory.</p>
<p>Eventually, I moved on from Leeds University to cofound Analyseit, and develop the successor to Astute. The ‘Module 1’ product transformed into the Clinical Laboratory Module and later into the Method Validation Edition. From its conception in Astute, to its implementation to Analyseit, the product has transformed method validation in clinical laboratories and invitro diagnostic companies around the world. </p>
<p>Without Rick, Analyseit would not exist today. Rick was the original catalyst and visionary for the product. Although I am writing this with great sadness, reflecting on the past, Rick’s enthusiasm for life leaves me positive and inspired. His inspiration will live on through Analyseit.</p>
<p>Simon Huntington</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/c59aNMMnJhc" height="1" width="1" alt=""/>
Method validation
Statistics
Wed, 29 Oct 2014 11:18:58 GMT
https://analyseit.com/blog/2014/10/asombrenoteprofessorrickjones

https://analyseit.com/blog/2014/8/analyseit380releasedprincipalcomponentanalysispca
Analyseit 3.80 released: Principal Component Analysis (PCA)
http://feeds.analyseit.com/~r/Analyseit/~3/THoUkTBdIj4/analyseit380releasedprincipalcomponentanalysispca
<div class="article">
<div class="content">
<p>Today we released version 3.80 of the Analyseit Standard edition.</p>
<p>The new release includes Principal Component Analysis (PCA), an extension to the multivariate analysis already available in Analyseit. It also includes probably the most advanced implementation of biplots available in any commercial package. </p>
<p>New features include:</p>
<ul>
<li>Principal Components Analysis (PCA). </li>
<li>Scree plot. </li>
<li>Classical Gabriel and modern Gower & Hand biplots. </li>
<li>Covariance and Correlation monoplots. </li>
<li>Color maps for correlations and patterned matrices, based on sign and magnitude, to help identify patterns. </li>
<li><a href="http://analyseit.com/docs/tutorials/correlation/overview" target="_blank">Tutorial (with video)</a> on how to visualize the relationships between variables and similarities between observations. </li>
</ul>
<p><img src="https://analyseit.com/assets/article/54611636/BiPlot.png"></p>
<p>The tutorial walks you through a guided example looking at how to use correlation and principal component analysis to discover the underlying relationships in data about New York Neighbourhoods. It demonstrates the amazing new features and helps you understand how to use them. You can either follow the tutorial yourself, at your own pace, or <a href="http://analyseit.com/docs/video/correlationpca" target="_blank">sit back and watch the video now</a>.</p>
<p><a href="http://analyseit.com/docs/video/correlationpca" target="_blank"><img src="https://analyseit.com/assets/article/54611636/CorrelationPCA.jpg"></a></p>
<p>If you have <a href="http://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="http://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired you can renew it to get this update and forthcoming updates, see <a href="http://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/THoUkTBdIj4" height="1" width="1" alt=""/>
Using Analyseit
In development
Releases
Statistics
Mon, 18 Aug 2014 12:46:05 GMT
https://analyseit.com/blog/2014/8/analyseit380releasedprincipalcomponentanalysispca

https://analyseit.com/blog/2014/6/recentimprovementsinanalyseit376andourfirstvideotutorial
Recent improvements in Analyseit 3.76 and our first video tutorial!
http://feeds.analyseit.com/~r/Analyseit/~3/VgL31Em788A/recentimprovementsinanalyseit376andourfirstvideotutorial
<div class="article">
<div class="content">
<p>If you <a href="https://www.facebook.com/analyse.it" target="_blank">follow us on Facebook</a> you will no doubt already know about the recent improvements in the Analyseit Method Validation edition and the release of our first video tutorial. If not, now is a good time to <a href="https://www.facebook.com/analyse.it" target="_blank">follow us</a> since we post short announcements and feature previews on Facebook, and use the blog only for news about major releases.</p>
<h4>What's new in release 3.76?<br>
</h4>
<p>The latest changes and improvements to the Analyseit Method Validation edition include:</p>
<ul>
<li>Extended BlandAltman Limits of Agreement (LoA) so you can calculate LoA using the ratio of measurements (for when difference between methods is related to magnitude of measurement), using linear regression (for when a transformation is not enough to eliminate the relationship between difference and magnitude), or using nonparametric percentiles (for when the differences are not normally distributed). </li>
<li>Added the <a href="http://analyseit.com/docs/userguide/methodcomparison/mountainplot" target="_blank">mountain plot</a> to method comparison so you can see the distribution of the differences. From the plot you can see the median of the differences, the central 95% interval, the range, and the percentage of observations outside the allowable error bands.</li>
<li>Added a new <a href="http://analyseit.com/docs/300/tutorials/blandaltman/overview" target="">indepth tutorial</a> to demonstrate how you can use Analyseit and the BlandAltman plot to determine the agreement between methods in various different scenarios. The tutorial is included in the software.</li>
</ul>
<p><img style="cursor: default;" src="https://analyseit.com/assets/article/54611635/BlandAltmanRegressionLoA.png"></p>
<p>If you have <a href="http://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you can download and install the update now, see <a href="http://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired you can renew it to get this update and forthcoming updates, see <a href="http://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.</p>
<h4>Watch our first video tutorial!<br>
</h4>
<p>Finally, we are delighted to release our first video tutorial. The tutorial is the video equivalent of the tutorial above. It walks and talks you through using Analyseit to determine the agreement between methods. Sit back and <a href="http://analyseit.com/docs/video/blandaltmanmethodagreement">watch the video now</a>. <br>
</p>
<p><a href="http://analyseit.com/docs/video/blandaltmanmethodagreement"><img style="cursor: default;" src="https://analyseit.com/assets/article/54611635/Intro1023.jpg"></a></p>
<p>We intend to produce more video tutorials in future, so let us know what you think: what you like, dislike, and how we can improve them in future.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/VgL31Em788A" height="1" width="1" alt=""/>
Method validation
Releases
Using Analyseit
Tue, 03 Jun 2014 13:15:29 GMT
https://analyseit.com/blog/2014/6/recentimprovementsinanalyseit376andourfirstvideotutorial

https://analyseit.com/blog/2014/2/analyseit370releasedimprovedmultiplecomparisonsconfidenceintervaldatabarsmeanmeanplotandsavefitmodelvariablesbacktothedataset
Analyseit 3.70 released: improved multiple comparisons, confidence interval data bars, meanmean plot, and save fit model variables back to the dataset
http://feeds.analyseit.com/~r/Analyseit/~3/Uuc34b9koXk/analyseit370releasedimprovedmultiplecomparisonsconfidenceintervaldatabarsmeanmeanplotandsavefitmodelvariablesbacktothedataset
<div class="article">
<div class="content">
<p>Today we released version 3.70 of Analyseit. <br>
</p>
<p>The new version includes many new features which some of you may have read about on our <a href="http://www.facebook.com/analyse.it" target="_blank">Facebook</a> page over the last few weeks:</p>
<p><img style="cursor: default;" src="https://analyseit.com/assets/article/54611634/MultipleComparisons.png"></p>
<p>New features include:<br>
</p>
<p>Compare groups</p>
<ul>
<li>Multiple comparisons: Hsu (with best), Scheffe (all contrasts), Steel (nonparametric against control), DwassSteelCritchlowFligner (nonparametric all pairs), Wilcoxon (individual comparisons).</li>
<li>pvalue on all multiple comparisons.</li>
<li>Confidence interval data bars on Excel 2010 and later.</li>
<li>MeanMean scatter plot for visualizing multiple comparisons. </li>
</ul>
<p>Fit Model</p>
<ul>
<li>Lack of fit test for all simple regression models: Line, Polynomials, Logarithmic, Exponential, Power.</li>
<li>Maximum attainable R2 in conjunction with lack of fit test.</li>
<li>Save variables back to the dataset: Fitted Y, Residuals, Standardized Residuals, Studentized Residuals, Leverage, Cook's Influence.</li>
<li>ttest on regression coefficients.</li>
<li>Standardized beta coefficients. </li>
</ul>
<p>Correlation</p>
<ul>
<li>Covariance matrix.</li>
<li>Correlation coefficient confidence interval data bars on Excel 2010 and later.</li>
</ul>
<p>If you have <a href="http://analyseit.com/docs/300/userguide/concepts/maintainance" target="_blank">active maintenance</a> you will be notified an update is available when you next start Analyseit, or you can download and install the update now, see <a href="http://analyseit.com/docs/300/userguide/concepts/softwareupdates" target="_blank">updating the software</a>. If maintenance on your licence has expired now is a good time to renew it to get this update and forthcoming updates, see <a href="http://analyseit.com/store/maintenance/" target="_blank">renew maintenance</a>.<br>
</p>
<p>We'll go into more detail on the new features over the next few weeks. In the meantime, <a href="http://www.facebook.com/analyse.it" target="_blank">like us on Facebook</a> to get the latest news, see announcements of forthcoming features, and get handy tips on how to use the myriad of features in Analyseit.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/Uuc34b9koXk" height="1" width="1" alt=""/>
Releases
Thu, 27 Feb 2014 13:34:49 GMT
https://analyseit.com/blog/2014/2/analyseit370releasedimprovedmultiplecomparisonsconfidenceintervaldatabarsmeanmeanplotandsavefitmodelvariablesbacktothedataset

https://analyseit.com/blog/2014/1/oursoftwaredevelopmentandvalidationprocess
Our software development and validation process
http://feeds.analyseit.com/~r/Analyseit/~3/eqkGc0mQpyg/oursoftwaredevelopmentandvalidationprocess
<div class="article">
<div class="content">
<p>Probably the greatest concern when using statistical software is reliability. Is the software producing accurate, numerically correct results that have been validated?</p>
<p>It’s a very important question. Many of you work in FDA and regulated environments where the penalties for mistakes are very high. And those of you outside such environments are still making important business and research decisions using Analyseit. It’s therefore imperative that the software you depend upon is developed to a professional standard, thoroughly tested and validated. </p>
<p>Today we are publishing details of the software development and validation process we use at Analyseit, see: </p>
<p><a target="_blank" href="https://analyseit.com/support/developmentandvalidationprocess">http://analyseit.com/support/developmentandvalidationprocess</a></p>
<p>For those of you working in regulated environments the document is essential to justify the use of offtheshelf, thirdparty software. </p>
<p>For everyone else it explains exactly how Analyseit is developed, tested and validated. You can use it as a checklist against which to judge the quality of alternatives. But be wary. Professional software development is complex and developing numerical software more so. We have over 25years software development experience (each developer, not aggregate), and over 20years developing statistical software. </p>
<p>For more about our track record, see:</p>
<p><a target="_blank" href="http://analyseit.com/company/aboutus">http://analyseit.com/company/aboutus</a></p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/eqkGc0mQpyg" height="1" width="1" alt=""/>
In development
Business
Releases
Tue, 14 Jan 2014 14:59:07 GMT
https://analyseit.com/blog/2014/1/oursoftwaredevelopmentandvalidationprocess

https://analyseit.com/blog/2013/12/version360nowincludesrepeatmeasuresanovaandfriedmantests
Version 3.60 now includes repeatmeasures ANOVA and Friedman tests
http://feeds.analyseit.com/~r/Analyseit/~3/GD2Ar09dPTk/version360nowincludesrepeatmeasuresanovaandfriedmantests
<div class="article">
<div class="content">
<p>We have just released version 3.60 of the Analyseit Standard edition. It now includes repeatmeasures ANOVA and Friedman tests in the Compare Pairs analysis. </p>
<p>If you have active maintenance, Analyseit will notify you an update is available in the next few days, or you can download it immediately at:</p>
<p><a href="http://analyseit.com/support/download">http://analyseit.com/support/download</a></p>
<p>If you do not have active maintenance, now is a great time to extend maintenance to get this latest update and all updates for the next 1 or 3years, see:</p>
<p><a href="http://analyseit.com/store/FAQ#maintenance">http://analyseit.com/store/FAQ#maintenance</a></p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/GD2Ar09dPTk" height="1" width="1" alt=""/>
Releases
Thu, 12 Dec 2013 15:42:01 GMT
https://analyseit.com/blog/2013/12/version360nowincludesrepeatmeasuresanovaandfriedmantests

https://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable
Analyseit Method Validation edition version 3.5 is now available
http://feeds.analyseit.com/~r/Analyseit/~3/peMElBEbpCI/analyseitmethodvalidationeditionversion35isnowavailable
<div class="article">
<div class="content">
<p>Today we released the Analyseit Method Validation edition version 3.5. The software is feature complete, validated, and includes documentation. It supports Excel 2007, Excel 2010 (32 and 64bit) and Excel 2013 (32 and 64bit). <br></p><p>We took this opportunity to rename the product from the Analyseit Method Evaluation edition to the Method Validation edition. The product is the same, but the new name better reflects the intended purpose of the product.</p><p><img style="cursor: default;" src="https://analyseit.com/assets/article/54611631/MethodValidation.PNG"></p>
<p>New features include:</p>
<p><b>Method Comparison</b></p>
<ul>
<li>All method comparison functions: scatters plots, difference plots, and fits, are now in one place so it’s easy to make decisions or choose fits after looking at the data.</li>
<li>Ordinary linear, Weighted linear, Ordinary Deming, Weighted Deming and Passing & Bablok regressions. </li>
<li>Constant difference and relative difference plots with mean difference, allowable difference, and limits of agreement.</li>
<li>Bootstrap confidence intervals for parameters and bias in Passing & Bablok regression.</li>
<li>Exact pvalues on CUSUM linearity test.</li>
<li>Predict mean difference at decision levels.</li>
<li>Equality and equivalence hypothesis tests at decision levels.</li>
<li>Kappa and Weighted Kappa qualitative method comparison.</li>
</ul>
<p><b>Precision</b></p>
<ul>
<li>GraybillWang / TingBurdickGraybillJeyaratnamLu modified large sample confidence interval estimators for variance components – produces confidence closer to the stated level than Satterthwaite confidence intervals, which can be too liberal.</li>
<li>Support for unbalanced 1 and 2factor nested random ANOVA models, for looking at precision between days, runs, laboratories, etc.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>HsiehLiu confidence interval estimator for degree of nonlinearity.</li>
<li>Equality and equivalence hypothesis tests for nonlinearity.</li>
</ul>
<p><b>Reference interval</b></p>
<ul>
<li>HarrellDavis, Bootstrap, and Robust Biweight quantiles.</li>
<li>BoxCox and other transformations.</li>
</ul>
<p><b>Diagnostic performance / ROC</b></p>
<ul>
<li>Compare up to 10 independent or correlated ROC curves. </li>
<li>Equality, equivalence and noninferiority hypothesis tests.</li>
<li>Estimate false positive fraction (FPF) at fixed sensitivity, sensitivity at fixed FPF, and sensitivity/FPF at fixed cutoff.</li>
<li>Find optimal decision threshold based on costs.</li>
</ul>
<p><b>Binary diagnostic tests</b></p>
<ul>
<li>Compare sensitivity and specificity of 2 independent or correlated tests.</li>
<li>Diagnostic oddsratio and Youden’s index.</li>
<li>Improved confidence interval estimators for likelihood ratio and predictive values.</li>
</ul>
<p>For more information about the new version, and to download a free 30day trial, see:</p>
<p><a href="http://analyseit.com/products/methodvalidation">http://analyseit.com/products/methodvalidation</a></p>
<p>Pricing for the Analyseit Method Validation edition starts at US$ 699 for a 1user perpetual licence. If you already have a licence you may qualify for a free upgrade, if you have active maintenance, otherwise you can extend maintenance to get the upgrade (and all updates for 1 or 3years) free of charge. To see if you qualify for a free upgrade, otherwise get a quote to extend maintenance, see:</p>
<p><a href="http://analyseit.com/support/download">http://analyseit.com/support/download</a></p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/peMElBEbpCI" height="1" width="1" alt=""/>
In development
Method validation
Releases
Mon, 14 Oct 2013 13:44:55 GMT
https://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable

https://analyseit.com/blog/2013/8/jointheanalyseitmethodevaluation35publicbetatestandearnafreeupgrade
Join the Analyseit Method Evaluation 3.5 public beta test and earn a free upgrade!
http://feeds.analyseit.com/~r/Analyseit/~3/B3mmPuozAgc/jointheanalyseitmethodevaluation35publicbetatestandearnafreeupgrade
<div class="article">
<div class="content">
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="http://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4>
<p>Today we released the first public beta test version of the Analyseit Method Evaluation edition, version 3.5. The software is feature complete and is validated – it is now only missing documentation.</p>
<p>We invite everyone to download the beta and try the new version of the software before it is finally released in September. You will need Excel 2007, 2010, or 2013 (32bit and 64bit versions are supported) and it can be installed and used alongside older versions of Analyseit so it won't interrupt your daytoday work.<br>
</p>
<p><img style="cursor: nwresize;" src="https://analyseit.com/assets/article/1/MethodComparisonScreenshot.png"></p>
<p>To download the beta version:</p>
<p><a target="_blank" href="http://analyseit.com/support/download/PTESTPKVAGWQKC7CS">http://analyseit.com/support/download/PTESTPKVAGWQKC7CS</a></p>
<p>To activate the software use the product key: </p>
<pre>PTESTPKVAGWQKC7CS</pre>
<h2>Will I qualify for a free upgrade?</h2>
<p>The software will be publically released at the end of September 2013. </p>
<p>If you purchased a licence in the last 12 months, the 12 months of maintenance included means you will qualify for a free upgrade to the new version. </p>
<p>If you are outside the 12 month free upgrade period you can purchase 12 months of maintenance, to get the upgrade (and all updates in the following year), for 20% of the cost of your licence. For example, if you have a 1user licence then the upgrade will cost 20% of the cost of a 1user licence. Similarly if you have a 3user licence the upgrade cost would be 20% of the cost of a 3user licence.</p>
<p>There is a way to reduce the cost…</p>
<h2>Earn a discount on your upgrade.</h2>
<p>The software is now validated but we would like your help to smoke out any differences in the statistics compared to Analyseit version 2.xx. To find differences, download the new release and compare the statistics against those shown by your current version of Analyseit. If you find a difference use the Send Feedback feature in the software to report the problem to us (see <a target="_blank" href="http://analyseit.com/docs/300/userguide/tasks/sendingfeedback">How to send feedback</a>).</p>
<p>If you find a difference in the statistics we will credit you $25 towards an upgrade (or new licence purchase), provided:</p>
<ul>
<li>The difference is not due to the use of a different algorithm or method. For example some statistics are now calculated using better formulas, so they will be numerically different.</li>
<li>The difference is not a small numerical difference due to a change in the order of calculation. Often a change in the order of calculation, or the splitting of computations between multiple processors, changes how rounding errors in floating point arithmetic accumulate, leading to a small difference.</li>
<li>Finally you must be using the latest release of the software. We will fix problems as they are reported and release an update that includes a fix, so please ensure you are using the latest version to avoid reporting problems that have already been fixed. Analyseit will automatically notify you of new releases, so ensure you install them as soon as possible.</li>
</ul>
<p>We will respond to every calculation problem reported. If it’s a genuine problem we will credit you US$ 25 that you can later use against the purchase of an upgrade or new licence. If it’s not a genuine problem we will tell you why (you will see the statistic does not change in later releases). There is no limit to the amount of credit you can earn, so please start testing now.</p>
<h2>Price increase from 1st September 2013.</h2>
<p>From the 1st September 2013 the price of the Analyseit Method Evaluation will increase slightly. A 1user perpetual licence will increase from US$ 649 to US$ 699, and a 1user concurrentuser perpetual licence will increase from US$ 1149 to US$ 1239. Pricing in other currencies will increase similarly.</p>
<p>The good news is you can buy a licence now, at the current prices, and the 1year maintenance included means you will get the new release (and all updates for the next year) free of charge. So if you’re considering buying a licence, or additional licences, buy now to beat the price increase.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/B3mmPuozAgc" height="1" width="1" alt=""/>
In development
Method validation
Releases
Wed, 14 Aug 2013 12:38:36 GMT
https://analyseit.com/blog/2013/8/jointheanalyseitmethodevaluation35publicbetatestandearnafreeupgrade

https://analyseit.com/blog/2013/6/analyseitmethodevaluationedition35alpha3nowavailable
Analyseit Method Evaluation Edition 3.5 alpha 3 now available
http://feeds.analyseit.com/~r/Analyseit/~3/TKr2YjWcsoU/analyseitmethodevaluationedition35alpha3nowavailable
<div class="article">
<div class="content">
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="http://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4><p>Today we released the 3<sup>rd</sup> alpha release of the Analyseit Method Evaluation Edition 3.5. Alpha releases are versions of the software that are still in active development, but are released to small group of customers so we can identify and fix any problems before the public beta release.</p>
<p>This release now completes the package with method comparison, which includes Deming regression, PassingBablok regression, and BlandAltman difference plots. Linearity, precision analysis, diagnostic performance (ROC analysis and binary test performance) and reference intervals were already included in earlier alpha releases.</p>
<p>If you would like to take part in this and subsequent test phases reply to this post or <a href="http://analyseit.com/company/contactus" target="_blank">contact us</a>. The test releases will run alongside any existing version of Analyseit, so your daytoday work won't be interrupted or affected. And those who help during testing will receive a discount on the upgrade (a free upgrade for those who contribute the most) when the product is released later this year.</p>
<p><img style="cursor: nwresize;" src="https://analyseit.com/assets/article/54611629/methcomp.PNG"></p>
<p><br>
</p><p>New features in the
release include:</p>
<p><b>Method Comparison</b></p>
<ul>
<li>All method comparison functions: scatters plots, difference plots, and fits, are now in one place so it’s easy to make decisions or choose fits after looking at the data.</li>
<li>Ordinary linear, Weighted linear, Ordinary Deming, Weighted Deming and Passing & Bablok regressions. </li>
<li>Constant difference and relative difference plots with mean difference, allowable difference, and limits of agreement.</li>
<li>Kappa and Weighted Kappa qualitative method comparison.<br>
</li>
<li>Bootstrap confidence intervals for parameters and bias in Passing & Bablok regression.</li>
<li>Exact pvalues on CUSUM linearity test.</li>
<li>Predict mean difference at decision levels.</li>
<li>Equality and equivalence hypothesis tests at decision levels.</li>
</ul>
<p><b>Precision</b></p>
<ul>
<li>GraybillWang / TingBurdickGraybillJeyaratnamLu modified large sample confidence interval estimators for variance components – produces confidence closer to the stated level than Satterthwaire confidence intervals, which can be too liberal.</li>
<li>Support for unbalanced 1way and 2way nested random ANOVA models.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>HsiehLiu confidence interval estimator for degree of nonlinearity.</li>
<li>Equality and equivalence hypothesis tests for nonlinearity.</li>
</ul>
<p><b>Reference interval</b></p>
<ul>
<li>HarrelDavis, Bootstrap, and Robust Biweight quantiles.</li>
<li>BoxCox and other transformations.</li>
</ul>
<p><b>ROC</b></p>
<ul>
<li>Compare up to 10 independent or correlated ROC curves. </li>
<li>Equality, equivalence and noninferiority hypothesis tests.</li>
<li>Estimate false positive fraction (FPF) at fixed sensitivity, sensitivity at fixed FPF, and sensitivity/FPF at fixed cutoff.</li>
<li>Find optimal decision threshold based on costs.</li>
</ul>
<p><b>Binary diagnostic tests</b></p>
<ul>
<li>Compare sensitivity and specificity of 2 independent or correlated tests.</li>
<li>Diagnostic oddsratio and Youden’s index.</li>
<li>Improved confidence interval estimators for likelihood ratio and predictive values.</li>
</ul>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
<a id="arss_hlComments" href="/blog/2013/6/analyseitmethodevaluationedition35alpha3nowavailable#comments">2 comments</a>
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/TKr2YjWcsoU" height="1" width="1" alt=""/>
Releases
Method validation
In development
Wed, 19 Jun 2013 11:29:37 GMT
https://analyseit.com/blog/2013/6/analyseitmethodevaluationedition35alpha3nowavailable

https://analyseit.com/blog/2013/5/analyseitmethodevaluationedition35alpha2nowavailable
Analyseit Method Evaluation Edition 3.5 alpha 2 now available
http://feeds.analyseit.com/~r/Analyseit/~3/cCzLkI3wJhM/analyseitmethodevaluationedition35alpha2nowavailable
<div class="article">
<div class="content">
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="http://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4><p>Today we released the 2<sup>nd</sup> alpha of the Analyseit Method Evaluation Edition 3.5. <br></p><p>Alpha releases are prerelease versions of the software that are still in active development. We release them to a small group of customers so we can get feedback and quickly identify and fix any problems before the public beta release. If you want to take part in the test phase reply or comment on to this post or <a href="http://analyseit.com/company/contactus" target="">contact us</a>. You can use prerelease versions of Analyseit alongside your
existing version of Analyseit, so it won't disrupt your work. And, if you help during in the test phases you will get a discount on the upgrade (a free upgrade for those who contribute the most) when the product is released later this year.</p>
<p>This latest alpha release includes linearity and precision analysis, plus diagnostic test performance (ROC analysis and binary test performance) and reference intervals from the 1<sup>st</sup> alpha. <br></p>
<p><img style="cursor: default;" src="https://analyseit.com/assets/article/54611628/LinearityDifferencePlot.png"></p>
<p><br>Some of the new features included so far are:</p><p><b>Precision</b></p>
<ul>
<li>GraybillWang / TingBurdickGraybillJeyaratnamLu modified large sample confidence interval estimators for variance components – produces confidence closer to the stated level than Satterthwaire confidence intervals, which can be too liberal.</li>
<li>Support for unbalanced 1way and 2way nested random ANOVA models.</li>
</ul>
<p><b>Linearity</b></p>
<ul>
<li>HsiehLiu confidence interval estimator for degree of nonlinearity.</li>
<li>Equality and equivalence hypothesis tests for nonlinearity.</li>
</ul>
<p><b>Reference interval</b></p>
<ul>
<li>HarrelDavis, Bootstrap, and Robust Biweight quantiles.</li>
<li>BoxCox and other transformations.</li>
</ul>
<p><b>ROC</b></p>
<ul>
<li>Compare up to 10 independent or correlated ROC curves. </li>
<li>Equality, equivalence and noninferiority hypothesis tests.</li>
<li>Estimate false positive fraction (FPF) at fixed sensitivity, sensitivity at fixed FPF, and sensitivity/FPF at fixed cutoff.</li>
<li>Find optimal decision threshold based on costs.</li>
</ul>
<p><b>Binary diagnostic tests</b></p>
<ul>
<li>Compare sensitivity and specificity of 2 independent or correlated tests.</li>
<li>Diagnostic oddsratio and Youden’s index.</li>
<li>Improved confidence interval estimators for likelihood ratio and predictive values.</li>
</ul>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
<a id="arss_hlComments" href="/blog/2013/5/analyseitmethodevaluationedition35alpha2nowavailable#comments">2 comments</a>
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/cCzLkI3wJhM" height="1" width="1" alt=""/>
In development
Method validation
Releases
Fri, 10 May 2013 14:45:46 GMT
https://analyseit.com/blog/2013/5/analyseitmethodevaluationedition35alpha2nowavailable

https://analyseit.com/blog/2013/4/announcingtheanalyseitmethodevaluationedition35
Announcing the Analyseit Method Evaluation Edition 3.5
http://feeds.analyseit.com/~r/Analyseit/~3/LlhdigApGyM/announcingtheanalyseitmethodevaluationedition35
<div class="article">
<div class="content">
<h4>Update 14th October 2013</h4>
<p>The Analyseit Method Validation Edition version 3.50 has now been released, see <a href="http://analyseit.com/blog/2013/10/analyseitmethodvalidationeditionversion35isnowavailable">Analyseit Method Validation edition version 3.5 is now available</a>.</p>
<h4>Original article</h4><p>We are now starting to release test previews of a major update to the Analyseit Method Evaluation edition. The new release will include many new features (we'll reveal more in the coming weeks) and will support 32 and 64bit versions of Excel 2007, 2010, and 2013.</p>
<p>During the initial test phases we release development versions of the application to a small group of customers to ensure it installs and runs as expected on a wide range of PCs and configurations. The official beta test phase stage then follows where more customers are invited to download and use the software, while we ironout the final few bugs before the official release. The official release is planned for summer 2013.</p>
<p><img style="cursor: default;" src="https://analyseit.com/assets/article/54611627/ReferenceIntervalPreview.png"></p>
<br>
<p><img style="cursor: default;" src="https://analyseit.com/assets/article/54611627/BinaryTestPerformancePreview.png"></p>
<p>If you want to take part in the test phase, reply to this post or <a href="https://analyseit.com/company/contactus">contact us</a> and let us know what aspects of Analyseit you use:</p>
<p>Analytical Linearity, Precision, Accuracy,<br>
Diagnostic performance (ROC, binary test performance),<br>
Reference ranges,<br>
Agreement (BlandAltman). </p>
<p>We will then invite you into the testing phase at the appropriate time over the next 46 weeks. You will be able to use the test version alongside your current Analyseit so it will not disrupt your daytoday work. </p>
<p>As a reward for your time, those who partipicate in the test phases will get a discount on the upgrade (or free upgrade for those who contribute the most) when the product is released later this year.</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/LlhdigApGyM" height="1" width="1" alt=""/>
In development
Releases
Method validation
Fri, 19 Apr 2013 15:39:36 GMT
https://analyseit.com/blog/2013/4/announcingtheanalyseitmethodevaluationedition35

https://analyseit.com/blog/2013/2/quantilespercentileswhysomanywaystocalculatethem
Quantiles, Percentiles: Why so many ways to calculate them?
http://feeds.analyseit.com/~r/Analyseit/~3/tPvk6BOW4aI/quantilespercentileswhysomanywaystocalculatethem
<div class="article">
<div class="content">
<p>What is a sample quantile or percentile? Take the 0.25 quantile (also known as the 25<sup>th</sup> percentile, or 1<sup>st</sup> quartile)  it defines the value (let’s call it <i>x</i>) for a random variable, such that the probability that a random observation of the variable is less than x is 0.25 (25% chance).</p>
<p>A simple question, with a simple definition? The problem is calculating quantiles. The formulas are simple enough, but a take a quick look on Wikipedia and you’ll see there are at least 9 alternative methods <a target="_blank" href="http://en.wikipedia.org/wiki/Quantile">http://en.wikipedia.org/wiki/Quantile</a>. Consequently, statistical packages use different formulas to calculate quantiles. And we're sometimes asked why the quantiles calculated by Analyseit sometimes don’t agree with Excel, SAS, or R.</p>
<h2>How are quantiles and percentiles calculated in Excel, SAS and R?</h2>
<p>Excel uses formula R7 (in the Wikipedia article) to calculate the QUARTILE and PERCENTILE functions. Excel 2010 introduced two new functions that use slightly different formulas, with different denominators: PERCENTILE.INC and PERCENTILE.EXC. </p>
<p>SAS, R and some other packages let you choose which formula is used to calculate the quantiles. While this provides some flexibility, as it lets you reproduce statistics calculated using another package, the options can be confusing. Most nonstatisticians don’t know when to use one method over another. When would you use the "Linear interpolation of the empirical distribution function" versus the "Linear interpolation of the modes for the order statistics for the uniform distribution on [0,1]" method? </p>
<h2>Why so many ways to calculate quantiles?</h2>
<p>Many of the formulas to calculate quantiles were developed when today's computing power wasn’t available. Believe it or not some are nearly 100 years old! Now they’re merely historical curiosities, but some remain in packages like SPSS that have their roots in the 1970s.</p>
<p>During the development of Analyseit we always ask: what’s the latest or best (in some sense) method we can use to calculate this statistic? </p>
<p>Filling Analyseit with all the formulas invented would be easy. But that doesn’t help the average user – users that don’t have the knowledge, or can’t invest the time needed to research the most suitable method. Instead we look at published research to find the best method. If there is no single best method we implement a small set of alternatives that provide the best solution in specific situations – situations we can clearly define and explain. </p>
<p>Every statistical test and estimator included in Analyseit has to pass this test to make the cut. We applied these principles when it came to quantiles.</p>
<p>Hyndman and Fan published a paper on calculating quantiles in 1996. It evaluated the methods used by popular statistics packages to calculate quantiles, with the intention to find a consensus on which all statistics packages could standardise. Of the 9 formulas used, 4 formulas satisfied five of the six properties desirable for a sample quantile and their derivations were deemed justified. Of those 4 formulas, Hyndman and Fan felt the "Linear interpolation of the approximate medians for order statistics" (method <a target="_blank" href="http://en.wikipedia.org/wiki/Quantile">R8 on the Wikipedia page</a>) formula was best due to the approximately medianunbiased estimates of the quantiles, regardless of the distribution. Of the remaining formulas, 2 were also distributionfree but were not unbiased, and the other was approximately unbiased only for the normal distribution. They concluded that formula R8 should be adopted as the standard across software packages. </p>
<p>That was 1996. Unfortunately little progress has since been made toward standardisation. Many statistical packages have a long history (even Analyseit is over 15years old now!) and most tend to stick to the same method to maintain backwards computability with older versions. Even R, a relative newcomer, doesn't use the recommended formula. It uses formula R7 by default, for compatibility with S (<a target="_blank" href="http://stat.ethz.ch/Rmanual/Rpatched/library/stats/html/quantile.html">http://stat.ethz.ch/Rmanual/Rpatched/library/stats/html/quantile.html</a>). Minitab, SPSS and SAS use R6.</p>
<h2>Consistency between the statistics and plots.</h2>
<p>To complicate the situation further quantiles and percentiles are also used in statistical plots. The Tukey boxplot, for example, uses the 1<sup>st</sup> and 3<sup>rd</sup> quartiles (0.25 and 0.75 quantiles) for the extent of the <i>box</i> element of the plot. </p>
<p>Frigge, Hoaglin and Iglewicz published a paper in 1989 that looked at how quantiles were calculated in 3 of the major statistical packages. They found that because each package used a different formula, each identified different observations as outliers. Confusing! They recommended statistical packages use the "Ideal or Machine Forths" formula for consistency, which is equivalent to using method R8 to calculate quartiles.</p>
<h2>What technique does Analyseit use?</h2>
<p>By now you can probably guess that we chose to use <a target="_blank" href="http://en.wikipedia.org/wiki/Quantile">R8 formula</a> in Analyseit.</p>
<p><br>
<img style="cursor: default;" alt="Analyseit uses the R8 formula (see Wikipedia) to calculate quantiles and percentile" src="https://analyseit.com/assets/article/54611626/Quantileformula.png"><br>
<br>
</p>
<p>Formula R8 is recommended as the standard in both papers cited above, for descriptive statistics and plots. And using a single formula avoids the confusing situation you’ll sometimes see with other statistics packages, where the 1<sup>st</sup> and 3<sup>rd</sup> quartiles used for the boxplot differ from sample .25 and .75 quantiles.</p>
<p>Of course using formula R8 can lead to some differences between the quantiles calculated by Analyseit and other packages, though often you'll only see it with small sample sizes. If possible, as you can in R, we recommend you change the quantile calculation to use the R8 formula. If not, you can be sure the statistics calculated by Analyseit are absolutely correct. And you can cite this article as to why!</p>
<p><b>Further reading:</b></p>
<p><a href="http://www.jstor.org/stable/2684934" target="_blank">Hyndman, R.J.; Fan, Y. (November 1996). "Sample Quantiles in Statistical Packages". The American Statistician 50 (4): 361–365. </a><br>
</p>
<p><a href="http://www.jstor.org/stable/2685173" target="_blank">Frigge, Michael; Hoaglin, David C.; Iglewicz, Boris (February 1989). "Some Implementations of the Boxplot". The American Statistician 43 (1): 50–54.</a><br>
</p>
</div>
<div id="arss_pnlLinks" class="links">
<div class="pullright">
<a id="arss_hlComments" href="/blog/2013/2/quantilespercentileswhysomanywaystocalculatethem#comments">2 comments</a>
</div>
</div>
</div><img src="http://feeds.feedburner.com/~r/Analyseit/~4/tPvk6BOW4aI" height="1" width="1" alt=""/>
Using Analyseit
In development
Statistics
Mon, 18 Feb 2013 14:39:54 GMT
https://analyseit.com/blog/2013/2/quantilespercentileswhysomanywaystocalculatethem