After multiple impute, it will generate several imptued dataset. Then we can use pool estimate to estimate the affection of missing data. but when I want to establish a predict data, I will devide the imputed data into train and test sets.
The question I want to ask is how to divide the all imputed data and then how to estimate the affection after I has devided all the imputed data into train and test sets?
Thanks! Who can give me some suggestions?
For cross-sectional data it works as follows:
1. Select (it could be random) observation i for the testing set, and use the remaining observations in the training set. Compute the error on the test observation.
2. Repeat the above step for i = 1, 2, … N-1, where N is the total number of observations.
3. Compute the forecast accuracy measures based on all errors obtained.
For time-series data, the procedure is sometimes known as a "rolling forecasting origin" because the "origin" (n+i-1) at which the forecast is based, rolls forward in time. It works as follows:
1. We select the observation at time (n+i) for the testing set and use the observations at times t= {1, 2, … (n+i-1)} to estimate the forecasting model. Then we compute the error on the forecast for the time (n+i).
2. The above step should be done for all i= {1, 2, … (T-n)}, where T is the total number of observations and the forecast error should be measured on each (n+i) period accordingly.
3. In the end, we compute the forecast accuracy measures based on all errors obtained.
I hope this is clear enough, but if you need more details, do not hesitate to ask.
Mihail Motzev , thanks for your answer. My data is cross-sectional. I will do the binary Logistic regression to explore the relationships of dependent variable and independent variables. But there are some missing data. I want to use mutilple imputation to impute the missing data. It has generated 5 imputed dataset.
I want to konw how to divide the all imputed data and then how to estimate the affection after I has devided all the imputed data into train and test sets?
Do you konw how to apply it in the R ( R Studio)?
Hey, sorry for the late answer, but I was traveling and didn't have secure Internet. I don't know how to use it with R because I use different software, which implement the concept of cross-validation automatically. The name is Insights and you can find it at knowledgeminer (dot) eu. You can download it and use it for 30 days free of charge and after that you can pay or not if you want to keep it and use it in the future.
- Badges
- Science topic