dear colleagues,Im trying to estimate random parameters model with Apollo, the code
### Clear memory
rm(list = ls())
### Load Apollo library
library(apollo)
library("readxl")
### Set working directory (only works in RStudio)
apollo_setWorkDir()
### Initialise code
apollo_initialise()
library("fastDummies")
# Step 2: Load data, detect cores, socioeconomic characteristics:
## Step 2.1: Load data:
database_orig
its actually simple, just do it simple:
lm(Y~Exp(-a)+b*c) the problem you need to read is about principal components. the reason is the model has more than two variables on the input...principanl components deduce which variables are significant to fit the output then with those two variables you do the limear regresion called lm(y~ax+bz+c) the thing is use eigen(X*Xt) to deduce which variables are more important in the regression. X is the matrix of observations called draws_... if i can add, i would say you are starting the carreer. when i was on first semesters i overcomplexed the calcs...good day :p
Alternatives to Apollo and Mixed Multinomial Logit (MMNL) estimation include Hierarchical Bayes (HB) estimation, Latent Class Models (LCM), or simpler Multinomial Logit (MNL) models when computational simplicity or limited data demands faster convergence and interpretability.
- Badges
- Science topic
- Similar topics
- Control Systems Engineering
- Filtering