Z-score standardizationI am a JMP Ver. 18 user at your company.
We are researching nutritional and immune scores as prognostic factors for malignant tumors.
For lung cancer surgery patients, data on preoperative C-CONUT scores, continuous variables of immunonutrition scores such as GNRI , NLR , Hb , and Plt , as well as data on patient prognosis (YOU) and survival days have been created in the statistical analysis software JMP . We are attempting to develop new markers to predict YOU using these variables.
- From Z- score standardization
- Research score（continuous）
IHPI = β1×C-CONUT + β2×GNRI + β3×log(NLR) + β4×Hb + β5×Plt + β6×Weight loss
We would like to create an Integrated Host Prognostic Index (IHPI) such as the following.
I'm trying to do this in JMP version 18 , but it's not working. First of all, step 1 doesn't work. Could you please tell me the specific steps to take?
Basic Data Analysis and Modeling
Re: Standardizing Z-scoresI may be misunderstanding your question, but if you want to standardize each column in your data table to a mean of 0 and a standard deviation of 1 (i.e., create a variable by subtracting the mean and dividing by the standard deviation), you can do so by following these steps:
(1) Select columns by their column headers in the data table. To select multiple columns, hold down the Shift or Ctrl key while selecting.
(2) Right-click and select [Create New Formula Column] → [Distribution] → [Standardization] from the menu that appears.
See also the following figure.
Note that the standardization performed in the above method is done column by column (which can cause discrepancies when, for example, listwise removing missing values in statistical analysis), and ignores row exclusion attributes.
