Simulate a Kiti-Cyprus-like farm dataset

Generates a synthetic cross-section that mimics the structure of the Koundouri & Nauges (2005) sample: 239 farms classified into vegetables (95), cereals (89), and citrus (55) producers, with inputs and revenues whose moments roughly match Table 1 of the paper. The crop-choice process is driven by environmental and farmer characteristics so the Heckman selection step has signal to recover.

Usage

simulate_kiti_data(n_total = 239, n_veg = 95, n_cer = 89, n_cit = 55, seed = 1)

Arguments

n_total

Integer. Total number of farms (default 239).

n_veg, n_cer, n_cit

Integer. Target counts per crop.

seed

Optional integer seed.

Value

A data frame with one row per farm and crop-choice dummies `vegetables`, `cereals`, `citrus`.

Details

This is for methodology demonstration only – it is not the original Cyprus data and is not intended to reproduce the paper's exact point estimates.

Examples

farms <- simulate_kiti_data(seed = 1)
table(farms$crop)
#> 
#>    cereals     citrus vegetables 
#>         89         55         95