# SPE: Empirical sorted partial effects (SPE) and inference In yuqimemeda/SortedEffects: Estimation and Inference Methods for Sorted Partial Effects and Classification Analysis

## Description

`SPE` conducts SPE estimation and inference at user-specifed quantile index. The bootstrap procedures follows algorithm 2.1 as in Chernozhukov, Fernandez-Val and Luo (2018). All estimates are bias-corrected and all confidence bands are monotonized. For graphical results, please use `SPEplot`.

## Usage

 ```1 2 3 4``` ```SPE(fm, data, method = "ols", var.type = "binary", var.T, compare, subgroup = NULL, samp_weight = NULL, us = c(1:9)/10, alpha = 0.1, taus = c(1:9)/10, B = 10, ncores = 1, seed = 1, bc = TRUE, boot.type = "nonpar") ```

## Arguments

 `fm` Regression formula. `data` Data in use. `method` Models to be used for estimating partial effects. Four options: `"logit"` (binary response), `"probit"` (binary response), `"ols"` (interactive linear with additive errors), `"QR"` (linear model with non-additive errors). Default is `"ols"`. `var.type` The type of parameter in interest. Three options: `"binary"`, `"categorical"`, `"continuous"`. Default is `"binary"`. `var.T` Variable T in interset. Should be a character type. `compare` If parameter in interest is categorical, then user needs to specify which two category to compare with. Should be a 1 by 2 character vector. For example, if the two levels to compare with is 1 and 3, then `c=("1", "3")`, which will calculate partial effect from 1 to 3. To use this option, users first need to specify var.T as a factor variable. `subgroup` Subgroup in interest. Default is `NULL`. Specifcation should be a logical variable. For example, suppose data contains indicator variable for women (female if 1, male if 0). If users are interested in women SPE, then users should specify `subgroup = data[, "female"] == 1`. `samp_weight` Sampling weight of data. If null then function implements empirical bootstrap. If data specifies sampling weight, the function implements weighted bootstrap. Input should be a n by 1 vector, where n denotes sample size. Default is `NULL`. `us` Percentile of interest for SPE. Should be a vector of values between 0 and 1. Default is `c(1:9)/10`. `alpha` Size for confidence interval. Shoule be between 0 and 1. Default is 0.1 `taus` Indexes for quantile regression. Default is `c(1:9)/10`. `B` Number of bootstrap draws. Default is set to be 10. For more accurate results, we recommend 500. `ncores` Number of cores for computation. Default is set to be 1. For large dataset, parallel computing is highly recommended since bootstrap is time-consuming. `seed` Pseudo-number generation for reproduction. Default is 1. `bc` Whether want the estimate to be bias-corrected. Default is `TRUE`. If `FALSE` uncorrected estimate and corresponding confidence bands will be reported. `boot.type` Type of bootstrap. Default is `boot.type = "nonpar"`, and the package implements nonparametric bootstrap. An alternative is `boot.type = "weighted"`, and the package implements weighted bootstrap.

## Value

The output is a list with 4 components: (1) `spe` stores spe estimates and confidence bounds; (2) `ape` stores ape estimates and confidence bounds; (3) `us` stores percentile index as in `SPE` command; (4) `alpha` stores significance level as in `SPE` command.

## Examples

 ```1 2 3 4 5``` ```data("mortgage") fm <- deny ~ black + p_irat + hse_inc + ccred + mcred + pubrec + ltv_med + ltv_high + denpmi + selfemp + single + hischl test <- SPE(fm = fm, data = mortgage, var.T = "black", method = "logit", us = c(1:9)/10) ```

yuqimemeda/SortedEffects documentation built on May 23, 2019, 9:51 a.m.