About Thierry_S

Thierry_S

Hi, I'm not able to reproduce your issue in JMP 14.1; it looks like you are using JMP 15 or above. I you rebuild your plot step by step, when does the label "Marker" appear? If possible, it would be helpful to see an example of your data table and graph to check if there are any customizations that are not visible on the plot you shared (e.g. row state in the table, or graphic elements in the plot). Also, I noticed that under the sub-menu "Points", you have the "Error Interval" checked as "Error Bar": is it possible to check that back to "Auto"? Let's see what we can do. Best, TS

Thierry_S

Hi, I don't think you need the list brackets { }. Best, TS

Thierry_S

Hi, Is it possible that the Save As command fails because a file with the same name exists in the target directory? If it is a possibility, you may want to test if the file exists (File Exist (path)) to let the user know. Just a thought. Best, TS

Thierry_S

Hi, Have you considered transposing your data (Rows become Column and reciprocaly)? You would then be able to use the Col Minimum and Col Maximum formulas. Best, TS

Thierry_S

Hi, From the look of the report image captured in your first post, you may have a problem with the distribution of the residuals. The model assumes that residuals should be distributed normally: in your case it looks rather asymmetrical but I cannot be sure without actual testing. To do that, the model output, choose Save Residuals, select this new column in your table and test for normality. From the look of it, you may want to try a modification of your your Total Citations with Log X or Log (X + 1) transformation. Considering that you have a very simple model i.e. 1 Y by 1 X, and that the distribution of Ys may be problematic, you may also look into a nonparametric test (e.g. Wilcoxon) in the Fit Y by X > One Way Analysis platform > Nonparametric These are my humble recommendations and input from more seasoned JMP users may be desirable. Best, TS

Thierry_S

Hi, I think there is a minor typo at the end of the script: in the last Match case, the receiving column should be :AgeYears[i], not :AgeMonths[i]. Best, TS

Thierry_S

Hi JMP Community, I found an ugly work around to reach my goal: 1) In the Discriminant Analysis platform > Canonical Plot: Click Customize 2) Copy the 3 non-editable graphic scripts for Normal 50% Contours, Means 95% CL Ellipses, and Group Labels into a new script window 3) Create the desired plot in Graph Builder (i.e. saved Canonical 1 vs. Canonical 2 values) 4) Click Customize and add the 3 saved graphic scripts describe above which our now editable (e.g. "Text Color", "Text Size", "Pen Size") I would love to hear the input from anybody with a better idea. Best, TS

Thierry_S

Sorry, I cannot do that. Best, TS

Thierry_S

Hi Ehud, I wonder if your Excel file is larger than 95 Mb which is the maximum attachment size allowed on the JMP Forum. Otherwise, If you use the "Reply" (not "Quick Reply"), you should see the following display at the bottom of the page: But I assume that is what you did in your prior attempts so there might be something wrong with your file and/or the JMP Forum. Best, TS

Thierry_S

Hi, I think you forgot to include the file(s) you referred to in your last message or there might be a problem with the JMP forum attachment module. Best, TS

Thierry_S

Hi JMP Community, I have conducted a Discriminant Analysis using measurements of 92 biomarkers associated with 3 biological states denoted as Dummy: B, C, and D; see canonical plot below: Now, after saving the Canonical Scores to by data table, I would like to fine tune this plot in GraphBuilder but I have not been able to figure out how to reproduce the Normal 50% Contours; here is what I have so far: the shaded ellipses are clearly not matching the original Canonical Plot above. Any help would be welcome. Best, TS

Thierry_S

Hi, One way to do this is 1) create a new window and a ListBox outside the loop, 2) Append each graph to the ListBox within the loop, and 3) Append the new window with the ListBox after the loop closes. I don't have the time to give you a complete script solution but you might be able to figure it out with the Scripting Help and Manual. Best, TS

Thierry_S

Hi Jim, Thanks for the solution, it works exactly as advertised with the Big Class example but it returns some unexpected results when I run it on my specific plot elements: I will have to look in more details for the right graphic element to capture after the unwanted elements are peeled off. Best, TS

Thierry_S

Hi JMP Community, I'm working on a simple JSL script that assembles multiple plots into a 6 x 5 grid. While I get acceptable results by capturing the plot area using the following line of code: img = gbr[FrameBox (1)] << get picture; where gbr is a Graph Builder report, I have not been able to find the right item to call in the Tree Structure to also capture the axis values (without the labels; see below an approximation of what I'd like to capture). Is there a direct capture method that I could use or do I need to capture the entire plot structure and "delete the unwanted elements? Or, is there a completely different method you would suggest? Thank you for your help. Best, TS

Thierry_S · ‎09-15-2020

Let's review: Log10 (7.5e-57) = Log10 (7.5) + Log10 (e-57) = Log10 (7.5) - 57 = 0.875 - 57 = -56.125

Thierry_S · ‎09-15-2020

Are you sure you used -Log10? Best, TS

Thierry_S · ‎09-15-2020

Hi JMP Community, With a bit of digging, I was able to find an Add-In developed by John Sall from JMP that calculates correctly the FDR p Value False Discovery Rate PValue Best, TS

Thierry_S · ‎09-14-2020

Hi, Based on this later post, I think we did not understand the statement of "same values"; we interpreted as column 1 = column 2 for each row but I surmise that you mean any column with identical values in each row (i.e. 0 StdDev). Is that correct? If it is, I or somebody more experience that I am should be able to provide a relatively short script. Best, TS

Thierry_S · ‎09-13-2020

Hi Yotam, That seems to be a perfect job for Graph Builder: 1) Drag your X column to the X Axis Box 2) Drag your Y Column to the Y Axis Box 3) Drag Column A to the Group X Box 4) Drag Column B to the Group Y Box Let me know it this works for you. Best, TS

Thierry_S · ‎09-12-2020

Hi JMP Community, For as long as I can remember, I have used an approximation of the Benjamini-Hochberg FDR calculation to account for multiple testing: > On a sorted list of Nominal P Values (smaller to larger) : BH p Val = Nom P Val * N Row()/Row () However, I'm aware of some limitation with this approach: For large Nominal p Values, the formula described above returns values > 1 (not by much but it tells me that something is not right) For Nominal p Values that are equal (or very close for one another), the formula returns slightly different values which I think should not be the case Hence, is there a better way to calculate the FDR? Thank you for your help. Best, TS

Thierry_S · ‎09-11-2020

Hi Gabriel, The beauty of JMP is that you can build relatively complex models to compare the effect of different factors on your outcome. Hence, I would recommend to you look into the Fit Model platform in the Help section. Specifically, the model would include the NDSI, the Genotype, and the interaction between those two factors: this interaction factor will allow you to compare the "slopes" of your regressions. I hope it helps. Best, TS

Thierry_S · ‎09-10-2020

Hi JMP Community, I need to estimate the probability of sampling different proportion of 100 subjects out of a 400 population belonging to 4 equally proportioned categories I used a simple binomial distribution approach but I'm not sure that it is the correct method. Here is the specifics about my questions Population: N = 400 Group A: 25% Group B: 25% Group C: 25% Group 😧 25% ==> what is the chance of having group A represented by less than 10 subject AND of having Group B represented by less than 10 patients in a sample of 100? The main issue that the actual sample number for one group influences the other groups: how do I account for that? Thanks for your help. Best, TS

Thierry_S · ‎09-07-2020

Hi, It really depends on the question you need to address (i.e. what is the hypothesis you want to test). At first glance, I would suggest that the type of grants should be a factor in your model (i.e. X variable) with the outcome (Y) being one of the continuous measures you listed (e.g. number of publications or number of citations). Considering that the amount of funding is also a property of the grants, I would include in as a factor too with a possible interaction between grant type and amount of funding. This will result in a 1- or 2-way ANOVA that should be pretty simple to interpret, but again: what is the question you need to address? Best, TS

Thierry_S · ‎09-04-2020

Hi, You can change the format of the Quantiles values by right-clicking on the table, choose Set Properties, and change the table display to fixed decimals. The resulting analysis display can then be captured as a Script with the format change included. Best, TS

Thierry_S · ‎09-04-2020

Hi Vinc, Just a final note: have you tried the "Control Chart Builder" which provides customization options akin to that available in Graph Builder? Best, TS

Thierry_S · ‎09-03-2020

Hi, I now understand what your problem is and I realized I don't know a work around to solve your issue. The Variability chart use a categorical structure for the X axis independent of the actual format of your data. Sorry for not providing a better answer. Best, TS

Thierry_S · ‎09-02-2020

Hi Vinc, Based on your original post, the Fit Y by X platform only lists the option of "Proportional X Axis" when the X-axis data is either ordinal or categorical. Hence, there might be some confusion about the actual properties of the column you use as X axis. Your values might be continuous but the column properties might be set up otherwise (e.g. if you have characters in the data)? When you go to Column Info menu (right-click on the column header), what is the Data Type and Modeling Type? Also under column properties, are there any entries such as Value Labels, Axis, or something else? Could you share the data table you are working on? If not, could you post an image of the Column Info panel for your X-Axis data column? Best, TS

Thierry_S · ‎09-02-2020

Hi Vincenzo, Without seeing what your data table looks like it is a bit difficult to address your question but I would look into the properties of the column you use as X variable to see if there are any special settings for the Axis category. Also, based on your description of the Fit Y by X platform, it looks like the column you use for X variable is ordinal or categorical. Hence, have you considered to switch to a numerical and continuous format (if it makes sense). Best, TS

Thierry_S · ‎08-28-2020

Hi, It looks like nobody wants to take a stab at your question, so let me share my humble opinion. First, it would be useful to understand your goal for this analysis: do you want to visualize the data and/or identify inherent structure in the data, or something else? Second, I think that you may be expecting too much of the clustering algorithms. In my mind, basic clustering is primarily a visualization method to group items based on their similarities; different methods provide different ways to evaluate and display similarities. Hence, the different "look" of the dendrograms according to the different methods. Third, if you are interested in identifying a classifier for your supplements based on amino acid composition, you may want to explore the Discriminant or the Principal Component platforms. Of note, I don't think that the size of the dataset affects the quality of the clustering per se: the more data you have add the more complex "branching" you will get. At a minimum, you need 3 items to run a clustering analysis: it may not provide deep insight but technically that is all you need. Best, TS

Thierry_S · ‎08-28-2020

Hi, I would like to help but the first section of script you posted does not make sense to me without more information: 1) You are defining dtSample twice using partially redundant methods 2) You Update dtSample with a dtSpecs that has not been defined (maybe elsewhere in you script?) 3) You are creating a new column in a table deftb that has not been defined (maybe elsewhere in you script?) 4) Your For Loop upper limits N Rows (dtRdr) use a third table that has not been defined either Hence, it is very difficult to provide an answer without having both the complete script and example input and output. Sorry for not being more constructive. Best TS

Date Registered	‎04-19-2018 01:39 PM
Date Last Visited	2 hours ago
Total Messages Posted	239
Total Kudos Received	120

Online Status	Online
Date Last Visited	2 hours ago