CESKalman: CESKalman

Description Usage Arguments Details Author(s) References Examples

View source: R/plot.CESKalman.R

The first plot contains the model fit of the relative quantities. The second is the process of relative augmenting technical change. The third plot is the demeaned-data series applied in estimation.

1	plot.CESKalman(Kalman,t0=1,tEnd=nrow(Kalman$data))

`Kalman`	an object of class CESKalman returned from the CESKalman function
`t0`	Start date. Only available for yearly data, else it is an index
`tEnd`	End date. Only available for yearly data, else it is an index

NOTE: As we are primarily interested to explain the model fit of the relative quantities (e.g. capital relative to labor), the estimated equation is first translated into quantities, denoted x_t: Δ x_{t}=α(x_{t-1}+σ p_{t-1}-μ_{t-1})+(κ_{i}-1)Δ p_{t}+ε_t. x_t is the relative quantities, p_t the relative prices (e.g. the user cost relative to the wage) and μ_{t} \equiv (σ-1)log(Γ_t) with Γ_t being the process of relative augmenting technical change (e.g. capital augmenting techncial change relative to labor augmenting technical change).

The first graph shows the values of x_{t}, the fitted values,\hat{x_{t}}=x_{t}-ε_t and the residuals ε_t. It is used to evaluate the fit of the model.

The second graph shows the process of relative augmenting technical change, i.e. log(Γ_t) and shows the direction of technical change. Of particular interest is if the direction has changed throught the sample and features so-called "medium run" fluctuations.

The last graph shows the data series of x_t-\bar{x} and -(p_t-\bar{p}) with \bar{x} and \bar{p} denoting the mean over time. Note that the relative price is plotted on the right axis and with a negative sign in front. If the two resulting series are possitively correlated, it implies that the long run elasticity is expected to be positive.

Christian Sandholm Kastrup <CST@dreamgruppen.dk>, Anders Farver Kronborg <ANK@dreamgruppen.dk> and Peter Philip Stephensen <PSP@dreamgruppen.dk>

Kronborg et al (2019) and Kastrup et al (2021)

## First, data is loaded with the Load_Data function (or any other data set)
data = Load_Data(Country="USA",tstart=1970,tend=2017)

data = cbind(data[,"q"],data[,"w"],data[,"K"],data[,"L"])

## The second step is a call to the CESKalman function
Kalman = CESKalman(data,grid.lambda=c(10,500,20),lambda_est_freely = T)

## Lastly we can plot the fit of the model:
plot(Kalman)