# iim: Make Indicator Matrices (Impulses, Steps, Trends) In gets: General-to-Specific (GETS) Modelling and Indicator Saturation Methods

## Description

Auxiliary functions to make, respectively, matrices of impulse indicators (`iim`), step indicators (`sim`) and trend indicators (`tim`)

## Usage

 ```1 2 3 4 5 6 7 8``` ```##make matrix of impulse indicators: iim(x, which.ones = NULL) ##make matrix of step indicators: sim(x, which.ones = NULL) ##make matrix of trend indicators: tim(x, which.ones = NULL, log.trend = FALSE) ```

## Arguments

 `x` either an integer (the length of the series in question) or a series (a vector or matrix) from which to use the time-series index to make indicators of `which.ones` the locations of the impulses. If NULL (the default), then all impulses are returned `log.trend` logical. If TRUE, then the natural log is applied on the trends

## Details

If `x` is a series or vector of observations, then the index of `x` will be used for the labelling of the impulses, and in the returned `zoo` object.

Note: For `sim` and `tim` the first indicator is removed, since it is exactly colinear with the others.

## Value

A `zoo` matrix containing the impulses

## Author(s)

Genaro Sucarrat, http://www.sucarrat.net/

`isat`, `zoo`

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14``` ```##generate series: y <- rnorm(40) ##make matrix of impulse indicators: mIIM <- iim(40) ##make matrix of step-indicators, but only every third: mSIM <- sim(y, which.ones=seq(1,40,3)) ##give quarterly time-series attributes to y-series: y <- zooreg(y, frequency=4, end=c(2015,4)) ##make matrix of trend-indicators with quarterly labels: mTIM <- tim(y) ```

gets documentation built on Jan. 7, 2018, 5:02 p.m.