Alan Roberts

About

Username
Alan Roberts
Joined
Visits
55
Last Active
Roles
Community Administrator

Comments

  • It will be a while before we are in a position to go live with this In the mean time, if you are familiar with VBA programming you could automate your analysis that way.... Alan
  • Hi No you are not missing anything. statistiXL currently only works by outputting a static set of results for a given set of data. We are currently looking at adding formulae to statistiXL to allow some automatic recalculations. These will nee…
  • This isn't a feature that we support in statistiXL but if you are OK with object orientated programming in VBA then you shouldn't have too much trouble working it out. Simply start a new VBA solution and add references to sxlmain.dll and whatever …
  • Thanks for the kind words. We are planning to add a module to handle missing values to Version 2 of statistiXL. We are just starting to work on V2 now and it should be out later this year. Cheers Alan
  • Thanks for the input, we will add your request to our Wish List and look into the practicalities of adding it. Alan
  • Thanks for the suggestion ... this one is already on our to-do list! Alan
  • Thanks for the input, we've added this to our list. Alan
  • P.S. a look through the samples in the helpf file included with statistiXL will show the output generated by statistiXL for Bivariate, Multivariate, Partial, Multiple and Canonical Correlations. Cheers Alan
  • Hi I'm not sure what you're asking here. Multiple correlation determines the correlation between each of a set of dependent variables and a group of independent variables. If you have 2 dependent variables (Y1 and Y2) and 3 independent variable…
  • Hi There is currently no way to enable macros. We are hoping to expose Excel Formulae for many of statistiXL's functions in a later version. For now you could write VBA code against the statistiXL DLLs though we are unable to provide technical s…
  • Fit tests provide the resultant p value for the analysis in the output (i.e. there is no percentage level that the tests are conducted at). Macro recording is not supported. Cheers Alan
  • Hi This analysis runs with the latest version of statistiXL (1.6 beta) referenced in my first reply at the top of this topic. Can you please download it and give it a go - it may have been updated since you last downloaded it? Cheers Alan
  • Hi xbenx I will have a look at this but given that you have more than 98% of your values in a single bin this isn't going to meet any theoretical distribution. Nearly all of your categories have 0 (or almost 0 ) readings. The dataset you have si…
  • Hi statistiXL currently only supports Linear Regression, not Exponential Regression. The only option open to you would be to transform your data to its linear equivalent. In this case I think you would need to take natural logs of both sides Ln…
  • Hi I would have alook through the statistiXL help file initially. For a good introduction and hand worked example see the following book... Zar, J. H. (1999). Biostatistical Analysis. 4th edition. Prentice-Hall, Englewood Cliff, New Jersey. Che…
  • Hi If you are wanting to include the labels in the analysis (i.e. as suggested by ticking the "Labels in First Row" option) then you need to include the labels (Row 1) in your range i.e. you should use Category Range: Sheet1!$A$1:$A$67 Freq Rang…
  • Hi Exactly what error do you get? I have installed the download I linked to on a clean PC and your data runs through it fine (though you will be asked to select the pool data option). Cheers Alan
  • Hi xbenx I think this is down to a bug that has already been addressed in Version 1.6 of statistiXL which will be released soon. You can download a beta version from here. Please uninstall your existing version before installing the new one. Le…
  • Hi For cluster analysis only Column 1 should contain labels (not row 1) and these will be the case labels (not variable labels). If you don't have case labels then keep the "Lables in First Colum" option unchecked and just select the dataset with…
  • If you can send the data to support@statistixl.com with an example of the settings you are using for the analysis and I will have a look and see why it is falling over. Alan
  • Hello tempuser. The link appears to be working now, can you please try again? Alan
  • Hello tempuser Thank you for your input. You are correct, statistiXL is incorrectly displaying the descriptive statistics for Set 2 (X) rather than Set 1 (Y) when running a partial or multiple correlation (while labeling them as the Set 1 descrip…
  • Hi Patrina Please see my reply to your other post. Cheers Alan
  • Hi Patrina If you send your dataset to support@statistixl.com I will have a look at it for you. Cheers Alan
  • Hi All help documentation is included in the statistiXL download. The Help File includes an overview of each statistical procedure plus step by step examples of each type of test. A reference list is included at the end of the Help File in order…
  • Hi Paul statistiXL does not require a balanced design (for most models) but it does require that there are no empty cells (except for models with no interaction terms and no nested factors). By this I mean that you can have a different number of…
  • Hi Volker Please see the email I have sent you. Cheers Alan
  • Hi Volker Your definitely seem to be having some unusual issues. I have just run the KW and Correlation stats on my system with disjoint ranges without issue. Using Example 2 on our Correlation spreadsheet (included with statistiXL) I entered th…
  • Hi Borriss Phil who is our statistician is away this week but I will ask him to have a look at your suggestions as soon as he returns. In brief however, I know we have discussed the issues of missing values before. While they are a common issue …
  • Hi Maisey Sorry for the delay in replying, my wife delivered our baby daughter on Monday/Tuesday so things have been a bit up-in-the-air here! If you can send me the spreadsheet you are having trouble with I will gladly have a look at it. Send i…