proc_SP500: Preprocess SP500 data

In bspcov: Bayesian Sparse Estimation of a Covariance Matrix

Preprocess SP500 data

Description

The proc_SP500 function preprocesses the SP500 stock data by calculating monthly returns for selected sectors and generating idiosyncratic errors.

Usage

proc_SP500(SP500data, sectors)

Arguments

SP500data	A data frame containing SP500 stock data with columns including: symbol Stock symbol. date Date of the stock data. adjusted Adjusted closing price of the stock. sector Sector of the stock.
sectors	A character vector specifying the sectors to include in the analysis.

Details

1. Calculates monthly returns for each stock in the specified sectors

2. Estimates the number of factors via hdbinseg::get.factor.model(ic="ah")

3. Uses POET::POET() to extract factor loadings/factors and form idiosyncratic errors

Value

A list with components:

Uhat	A matrix of idiosyncratic errors.
Khat	Estimated number of factors.
factorparthat	Estimated factor returns.
sectornames	Sector for each column of Uhat.

Examples

data("SP500")
set.seed(1234)
sectors <- c("Energy", "Financials", "Materials",
   "Real Estate", "Utilities", "Information Technology")

Uhat <- proc_SP500(SP500, sectors)$Uhat
PPPres <- thresPPP(Uhat, eps = 0, thres = list(value = 0.0020, fun = "hard"), nsample = 100)
postmean <- estimate(PPPres)
diag(postmean) <- 0 # hide color for diagonal
plot(postmean)

bspcov documentation built on July 3, 2025, 1:10 a.m.