Hamilton_table_2

Standard deviation of cyclical component and correlation with cyclical
component of GDP for assorted macroeconomic series.

datasets

In the working paper titled "Why You Should Never Use the Hodrick-Prescott Filter", James D. Hamilton proposes a new alternative to economic time series filtering. The neverhpfilter package provides functions and data for reproducing his work. Hamilton (2017) <doi:10.3386/w23429>.

Justin M. Shea

neverhpfilter

An Alternative to the Hodrick-Prescott Filter

Hamilton_table_2 function

A <code>data.frame</code> containing 13 economic time series observations
of 5 variables:<dl>
 <dt>**cycle.sd:**</dt>
<dd>Standard deviation of the cycle component,
 computed with function <code>yth_filter(x, output = "cycle")</code>.</dd> <dt>**gdp.cor:**</dt>
<dd>Correlation of <code>cycle.sd</code> with the cycle.sd of
 <code>100 * log(RealGDP)</code>.</dd> <dt>**random.sd:**</dt>
<dd>Standard deviation of a random walk, computed by
 rolling differencing the series by period <code>h</code>, the same <code>h</code> passed
 to <code>yth_filter(x, output = "cycle")</code> to compute <code>cycle.sd</code>.</dd> <dt>**gdp.rand.cor:**</dt>
<dd>Correlation of <code>random.sd</code> with the random.sd
 of <code>100 * log(RealGDP)</code>.</dd>
</dl>

Format

Filtered series were based on the full sample available for that variable,
while correlations were calculated using the subsample of overlapping values
for the two indicators. Note that the regression residuals lose the first 11
observations and the random-walk calculations lose the first 8 observations.

Notes

Hamilton_table_2 — Hamilton_table_2

A <code>data.frame</code> containing 13 economic time series observations
of 5 variables:<dl>
 <dt>**cycle.sd:**</dt>
<dd>Standard deviation of the cycle component,
 computed with function <code>yth_filter(x, output = "cycle")</code>.</dd>

 <dt>**gdp.cor:**</dt>
<dd>Correlation of <code>cycle.sd</code> with the cycle.sd of
 <code>100 * log(RealGDP)</code>.</dd>

 <dt>**random.sd:**</dt>
<dd>Standard deviation of a random walk, computed by
 rolling differencing the series by period <code>h</code>, the same <code>h</code> passed
 to <code>yth_filter(x, output = "cycle")</code> to compute <code>cycle.sd</code>.</dd>

 <dt>**gdp.rand.cor:**</dt>
<dd>Correlation of <code>random.sd</code> with the random.sd
 of <code>100 * log(RealGDP)</code>.</dd>


</dl>

Filtered series were based on the full sample available for that variable,
while correlations were calculated using the subsample of overlapping values
for the two indicators. Note that the regression residuals lose the first 11
observations and the random-walk calculations lose the first 8 observations.

Hamilton_table_2: Hamilton_table_2

Description

Usage

Arguments

Format

Notes

Examples