weights_exponential

a single <code>numeric</code> to indicate the lag length.

a <code>numeric</code> vector of decay factors, between 0 and 1, but multiplied by 10 in
the implementation.

alphas

<code>TRUE</code> if the inverse exponential curves should be calculated as well.

do.inverse

a <code>logical</code>, if <code>TRUE</code> weights are normalized to unity.

do.normalize

Computes exponential weighting curves. Handy to self-select specific time aggregation weighting schemes
for input in <code>ctr_agg</code> using the <code>weights</code> argument.

Optimized prediction based on textual sentiment, accounting for the intrinsic challenge that sentiment can be computed and pooled across texts and time in various ways. See Ardia et al. (2021) <doi:10.18637/jss.v099.i02>.

Samuel Borms

sentometrics

An Integrated Framework for Textual Sentiment Time Series
Aggregation and Prediction

David Ardia

Keven Bluteau

Kris Boudt

Jeroen Van Pelt

Andres Algaba

weights_exponential function

<dl><dt>n</dt>
<dd>a single <code>numeric</code> to indicate the lag length.</dd>
<dt>alphas</dt>
<dd>a <code>numeric</code> vector of decay factors, between 0 and 1, but multiplied by 10 in
the implementation.</dd>
<dt>do.inverse</dt>
<dd><code>TRUE</code> if the inverse exponential curves should be calculated as well.</dd>
<dt>do.normalize</dt>
<dd>a <code>logical</code>, if <code>TRUE</code> weights are normalized to unity.</dd></dl>

Arguments

Compute exponential weighting curves — weights_exponential

<dl>

<dt>n</dt>
<dd>a single <code>numeric</code> to indicate the lag length.</dd>


<dt>alphas</dt>
<dd>a <code>numeric</code> vector of decay factors, between 0 and 1, but multiplied by 10 in
the implementation.</dd>


<dt>do.inverse</dt>
<dd><code>TRUE</code> if the inverse exponential curves should be calculated as well.</dd>


<dt>do.normalize</dt>
<dd>a <code>logical</code>, if <code>TRUE</code> weights are normalized to unity.</dd>

</dl>

weights_exponential: Compute exponential weighting curves

Description

Usage

Value

Arguments

See Also