Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/econometrics-r
R-based econometric analysis for academic research.
Passed 22 of 23 test cases with the skill, and 16 of 23 without it.
and 5 more in the full report
6 fixed16 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
When training random forests or non-parametric ML models for causal inference in R, what command should always be run beforehand to guarantee exact numerical reproducibility?
| Checked | Without skill | With skill |
|---|---|---|
| Executes set.seed() before model training. | Missed | Passed |
The instructions your model reads when the skill loads.
library(data.table) # Data manipulation
library(fixest) # Fixed effects estimation
library(modelsummary) # Regression tables
library(ggplot2) # Visualization
library(sf) # Spatial data
library(here) # Project paths
# Read and assign
dt <- fread(here("data", "raw", "file.csv"))
# Common operations
dt[, new_var := old_var * 100] # Create variable
dt[, mean_y := mean(y, na.rm = TRUE), by = group] # Group operations
dt[year >= 2000 & treated == 1] # Filter
dt[, .(mean_y = mean(y), n = .N), by = group] # Summarize
dt[other_dt, on = .(id, year)] # Merge
# Lag/lead within groups
setorder(dt, id, year)
dt[, lag_y := shift(y, 1), by = id]
dt[, lead_y := shift(y, -1), by = id]
# Two-way fixed effects
est1 <- feols(y ~ treatment + controls | id + year, data = dt)
# Clustered standard errors (default: fixed effect groups)
est2 <- feols(y ~ treatment | id + year, data = dt, cluster = ~state)
test cases passed, vs 16 / 23 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.