New R Code
Budget: €30 – €250 EUR
Good afternoon,
I am a University Professor of Psychology and I have a project funded by FCT in the field of criminology. I would like to develop software (which I have partially programmed in Excel already). This software aims to: 1) work with dichotomous variables that are temporally distinct (for example, at 3 or 4 fixed and different times); 2) map all the sequences that appear in the sample more often than expected by chance, i.e., more than the upper CI of the multiplication of their individual probabilities (each variable at each time would generate a different sequence), and then assess each sequence's probability of predicting a dichotomous event, calculated via ORs.
From the list of variables that appeared first in time, let's say, the first assessment was at age 8, to perform a calculation of the Median OR (and 75/25, 95/5 CIs, and max and min) for all the predictive assessments including the variable X at the origin of the sequence (to run that for each sequence).
The same for the last variable in the sequence.
Currently, TraMineR package in R allows mapping dichotomous variables and performing cross-tabs like Chi, but it does not compute odds ratios, nor calculates Median OR for all the variables starting with X. What would the budget be and how long would it take to do this in R or another platform? I have programmed almost everything that R does not do in Excel, so it would probably be replicating the code in R.
Thank you,
Miguel
I am a University Professor of Psychology and I have a project funded by FCT in the field of criminology. I would like to develop software (which I have partially programmed in Excel already). This software aims to: 1) work with dichotomous variables that are temporally distinct (for example, at 3 or 4 fixed and different times); 2) map all the sequences that appear in the sample more often than expected by chance, i.e., more than the upper CI of the multiplication of their individual probabilities (each variable at each time would generate a different sequence), and then assess each sequence's probability of predicting a dichotomous event, calculated via ORs.
From the list of variables that appeared first in time, let's say, the first assessment was at age 8, to perform a calculation of the Median OR (and 75/25, 95/5 CIs, and max and min) for all the predictive assessments including the variable X at the origin of the sequence (to run that for each sequence).
The same for the last variable in the sequence.
Currently, TraMineR package in R allows mapping dichotomous variables and performing cross-tabs like Chi, but it does not compute odds ratios, nor calculates Median OR for all the variables starting with X. What would the budget be and how long would it take to do this in R or another platform? I have programmed almost everything that R does not do in Excel, so it would probably be replicating the code in R.
Thank you,
Miguel