hepi: Bilateral Hedonic Elementary Price Indices
In hepi: Functions for estimating hedonic elementary price indices

Description Usage Arguments Details Value Author(s) References

View source: R/hepi.R

This function estimates bilateral hedonic elementary price indices based on estimations of the hedonic function in the base and the current period as well as a reference sample of quality characteristics.

hepi(hf0, hf1, M, 
     type = c("jevons", "dutot", "carli", "hdutot", "hcarli"), 
     na.rm = TRUE, debug = FALSE)

hepi.jevons(hf0, hf1, M)
hepi.carli(hf0, hf1, M)
hepi.dutot(hf0, hf1, M)
hepi.hcarli(hf0, hf1, M)
hepi.hdutot(hf0, hf1, M)

`hf0`	The base period hedonic function estimate. Must be of class `"hedonic.function"`.
`hf1`	The current period hedonic function estimate. Must be of class `"hedonic.function"`.
`M`	The reference sample.
`type`	The type of the index estimator(s) to be used. Can be a vector if estimates of several types are requested.
`na.rm`	A logical value indicating whether observations containing `NA` values should be stripped before the computation proceeds.
`debug`	A logical value indicating whether predicted prices should be returned for debugging purposes.

This function yields an estimate of a bilateral hedonic price index. Inputs are the two estimated hedonic functions hf0 and hf1 of the base and current period respectively. Both of these must be of class "hedonic.function". (See hedonic.function for a constructor of a hedonic.function object.)

The third input is the reference sample M to be used for the estimation of the index. This is usually a data frame containing N characteristics vectors to which both hedonic functions are applicable.

The type argument lets one choose the index formula to be used (and yet the index to be estimated). Currently, we implemented five alternative estimators, namely the

Jevons

Dutot

Carli

Harmonic Dutot and

Harmonic Carli

formulae. Details can be found in Chapter 6 of the reference mentioned below.

If debug == FALSE, this function returns a vector with the same length as type containing the index estimates. They are returned in the same order as given by type.

If debug == TRUE, this function returns a list with the following entries

`index`	The vector of index estimates as above.
`p0hat`	The vector of predicted prices p^0=h^0(M) in the base period.
`p1hat`	The vector of predicted prices p^1=h^1(M) in the current period.
`ratios`	The vector of price ratios p_n^1/p_n^0 (n=1,…,N).

Michael Beer r-hepi@michael.beer.name

Beer, M. (2007) Hedonic Elementary Price Indices: Axiomatic Foundation and Estimation Techniques. PhD thesis, University of Fribourg Switzerland, http://www.michael.beer.name/phdthesis.

hepi documentation built on May 2, 2019, 6:17 p.m.