Alan Roberts

About

Username
Alan Roberts
Joined
Visits
55
Last Active
Roles
Community Administrator

Comments

  • Hi Unfortunately the current version of statistiXL doesn't include CHAID or CART. We will add them to our wish list. Best Regards Alan
  • Hi seabc I hope this addresses your question. Please let me know if we have misunderstood you! The current version of statistical does not support k-means clustering, but this is an option that we are considering adding in the future. However, we…
  • Hi Mike What versions of statistiXL, Excel and Windows are you using? What errors are you receiving? Do you have any other add-ins installed in Excel? Are your copies of Windows and Excel up-to-date with regards all of the latest patches from M…
  • Hi There is no special limit at 62 rows and I have just run a 72 row x 3 column test without issue. If you would like to post your dataset to support@statistixl.com I will have a look at it for you. Cheers Alan
  • Hi Mercedes Yes, to calculate the correct scores you simply have to standardise the raw values for the variables before entering them into the equation. When working with a PCA based on a correlation matrix you simply standardise the raw values s…
    in PCA Comment by Alan Roberts May 2005
  • Hi Klaus I have forwarded your query on to Philip and he should have a reply for you tomorrow. Best Regards Alan
  • Hi Size shouldn't matter :-). If you want to send a sample dataset to support@statistixl.com I will have a look through it and see why it is failing. Best Regards Alan
  • Hi Generally statistiXL performs most tests in only a matter of seconds (or less). The clustering algorithm is quite CPU intensive and a large dataset could, however, result in a prolonged analysis. It is very unlikely that statistiXL has crash…
  • Hi Robertovi I have looked at your data and the reason the analysis fails is that you have too few residual degrees of freedom. Basically you have too many repeats and not enough cases. I have adjusted statistiXL so that in the next version it w…
  • Hi James Please see Phil's reply to your other posting on this topic. Cheers Alan
  • Hi James The trial version is fully functional. Are you saying you have 2 nexted factors of 2 standard factors plus a nested factor? Cheers Alan
  • Hi I'm not sure exactly what your problem is. Are you saying that you can run tests using the sample data but not your own data? If so, can you please email me one of the datasets that you are having problems with and screen shots of the exact s…
  • Hi This is not an issue I have come across before. statistiXL does not manipulate the spreadsheet other than to output the results so it should not be causing recalculation events to fire. Can you please provide me with some details on your syst…
  • Hi Unfortunately, statistiXL does not currently support 2 levels of repeats, although it can deal with multiple levels of a non-repeat factor (see examples 13 and 14 in the ANOVA section of the statistiXL Help File). In the example you cite, we c…
  • Hi Kennic There are many different ways to calculate percentiles. The well know main-frame descended stats package SAS for instance provides 5 different ways for calculating them. Your values are indeed appropriate for one such measure. The for…
  • Hi I am unable to reproduce the error you are reporting. With the data you have supplied starting at cell A1 (top left) I am entering the following Category Range: Sheet1!$A$1:$A$10 Frequency Range: Sheet1!$B$1:$B$10 Output Range: Sheet1!$D$1 T…
  • Yes, press F2
  • Yes and No . Most of the statistiXL number crunching is performed in a bunch of DLLs. You can see these in your statistiXL program directory. They all start with the prefix SXL and the rest of the name typically describes the module they rela…
  • Hi Peter You can enter your data in 2 ways. One way is in the format you list, with Set 1...Set 7 in the first row and the data underneath. If we start at cell A1, the sample you have provided would extend from A1 to G5. In this case, if your d…
  • Hi Miika It's hard for me to answer without knowing more about your data. I would suggest that you have a read through the statistiXL help file, looking especially at the examples for ANOVA to see if your data matches the sort used there. To me …
  • Hi Miner If you send a copy of a sheet you have trouble with to support@statistixl.com I will have a look at it and try to see what is going on. Cheers Alan
  • I have not heard of this error before? Is it only when working with a specific workbook or with different books? Does it happen for all analyses or just clustering? If you can send me a sample dataset I will see if I can reproduce the issue. Ch…
  • Hi Matt Unfortunately Philip who is the primary statistician for statistiXL is away at a conference until the end of next week. If you can send a sample dataset to me though (alan@statistixl.com) I will have him check through it as soon as he get…
  • Hello Daniel Unfortunately statistiXL will not do what you require. You can download a trial of statistiXL at http://www.statistixl.com to see the features that it has but your key number scenario is not covered. Best regards Alan Roberts stati…
  • Thanks for the comments. We do hope to expand the multivariate routines in the future to include Correspondence Analysis and other useful procedures, though we don't have a time scale for this yet. We are currently working on expanding the regres…
  • This has now been implemented and will be available in Version 1.5. If anyone needs access to this feature before the release of version 1.5 then please email a request to support@statistixl.com.
  • Hi Can you please let me know what Version of Windows you are using. Also, what language versions of Excel and Windows you are using (eg US English etc). Finally, there should be a file called statistixl.log located in the folder in which you in…
  • Can you send a sample dataset to support@statistixl.com for us to look at? Cheers Alan
  • We will look at adding this in time for the next release (Version 1.5) Cheers Alan statistiXL
  • The 1000 case limit has been removed in Version 1.4 Cheers Alan