gatelyPriceAsymm: Decompose Prices

In efriedland/friedland: Making Functions Do Your Econometrics For You

Description

Produces a matrix of 3 variables decomposed from the input time series variable. The historical maximum at each date, the cumulative positive changes, and the cumulative negative changes.

Usage

gatelyPriceAsymm(x, prefix = "P")

Arguments

x	The Time Series variable to be decomposed.

(prefix) The string that leads "max" "rec" and "cut" which is by default "P"

References

Gately (1991) "Imperfect Price-Reversibility of U.S. Gasoline Demand: Asymmetric Responses to Price Increases and Declines.

See Also

Asymmetry

Examples

# Data
Asymmetry.ts <- ts(Asymmetry[,-1], start = c(2003,6), frequency = 12)
decomposed <- gatelyPriceAsymm(Asymmetry.ts[,"Pcrude"], prefix = "Pcrude_")
newdat.ts <- ts.intersect(Asymmetry.ts, decomposed)
colnames(newdat.ts) <- c(colnames(Asymmetry.ts), colnames(decomposed))
head(newdat.ts)

# 2 models
lessrestricted <- dynlm(Pgasoline ~ Pcrude_max + Pcrude_rec + Pcrude_cut, newdat.ts)
summary(lessrestricted)
restricted <- dynlm(Pgasoline ~ Pcrude, newdat.ts, 
                    start = start(lessrestricted)) # fix the dates to be equal
summary(restricted)

# Choosing between them
anova(dynlm(Pgasoline ~ Pcrude_max + Pcrude_rec + Pcrude_cut, newdat.ts), 
      dynlm(Pgasoline ~ Pcrude, newdat.ts, start = c(2003,7)))
# F test shows significant difference. Use less restricted with smaller RSS. 

efriedland/friedland documentation built on Feb. 9, 2021, 11:53 a.m.