Learn R Programming

FLAME (version 1.0.0)

FLAME_PostgreSQL: PostgreSQL Database Implementation

Description

FLAME_PostgreSQL applies the FLAME algorithm based on PostgreSQL. If your computer system does not have PostgreSQL installed, install from here. For setup of PostgreSQL server, please refer to this tutorial. User must connect to PostgreSQL server in R using the command dbConnect(dbDriver('PostgreSQL'), dbname="your_dbname", host='your_localhost', port='your_port', user='your_username', password = 'your_password') and name the connection as db

Usage

FLAME_PostgreSQL(db, data, holdout, compute_var = FALSE,
  tradeoff = 0.1, PE_function = NULL, model = NULL,
  ridge_reg = NULL, lasso_reg = NULL, tree_depth = NULL)

Arguments

db

name of the database connection (must name the connection as db)

data

input data

holdout

holdout training data

compute_var

indicator variable of computing variance (optional, default = FALSE)

tradeoff

tradeoff parameter to compute Match Quality (optional, default = 0.1)

PE_function

user defined function to compute predictive error (optional)

model

user defined model - Linear, Ridge, Lasso, or DecisionTree (optional)

ridge_reg

L2 regularization parameter if model = Ridge (optional)

lasso_reg

L1 regularization parameter if model = Lasso (optional)

tree_depth

maximum depth of decision tree if model = DecisionTree (optional)

Value

(1) list of covariates FLAME performs matching at each iteration, (2) Sizes, conditional average treatment effects (CATEs), and variance (if compute_var = TRUE) of matches at each iteration, (3) match quality at each iteration, and (4) the original data with additional column *matched*, indicating the number of covariates each unit is matched on. If a unit is never matched, then *matched* will be 0.

Examples

Run this code
# NOT RUN {
data <- data(toy_data)

drv <- dbDriver('PostgreSQL')

db <- dbConnect(drv, dbname="FLAME", host='localhost',
port=5432, user="postgres", password = 'new_password')

FLAME_PostgreSQL(db = db, data = data, holdout = data)

dbDisconnect(db)
# }

Run the code above in your browser using DataLab