predict.sentomodel

a <code>sentomodel</code> object created with <code><a rd-options="" href="/link/sento_model?package=sentometrics&version=0.2" data-mini-rdoc="sentometrics::sento_model">sento_model</a></code>.

object

a <code>matrix</code> of <code>numeric</code> values with all explanatory variables to be used for the prediction(s),
structured row-by-row; see documentation for <code><a rd-options="" href="/link/predict.glmnet?package=sentometrics&version=0.2" data-mini-rdoc="sentometrics::predict.glmnet">predict.glmnet</a></code>. The number of variables should be equal to
<code>sentomodel$nVar</code>, being the sum of the number of original sentiment measures and the number of additional explanatory
variables. Variables discarded in the regression process are discarded again here, based on <code>sentomodel$discarded</code>.

newx

type of prediction required, a value from <code>c("link", "response", "class")</code>, see documentation for
<code><a rd-options="" href="/link/predict.glmnet?package=sentometrics&version=0.2" data-mini-rdoc="sentometrics::predict.glmnet">predict.glmnet</a></code>.

type

not used. Any values here will be ignored.

offset

Prediction method for <code>sentomodel</code> class, with usage along the lines of
<code>predict.glmnet</code>, but simplified in terms of allowed parameters.

Time series analysis based on textual sentiment, accounting for the intrinsic challenge that sentiment can be computed and pooled across texts and time in many ways. As described in Ardia et al. (2017) <https://ssrn.com/abstract=3067734>, the package provides a means to model the impact of sentiment in texts on a target variable, by first computing a wide range of textual sentiment measures and then selecting those that are most informative.

Samuel Borms

sentometrics

An Integrated Framework for Textual Sentiment Time Series
Aggregation and Prediction

David Ardia

Keven Bluteau

Kris Boudt

predict.sentomodel function

a <code>sentomodel</code> object created with <code><a rd-options='' href='sento_model'>sento_model</a></code>.

a <code>matrix</code> of <code>numeric</code> values with all explanatory variables to be used for the prediction(s),
structured row-by-row; see documentation for <code><a rd-options='' href='predict.glmnet'>predict.glmnet</a></code>. The number of variables should be equal to
<code>sentomodel$nVar</code>, being the sum of the number of original sentiment measures and the number of additional explanatory
variables. Variables discarded in the regression process are discarded again here, based on <code>sentomodel$discarded</code>.

type of prediction required, a value from <code>c("link", "response", "class")</code>, see documentation for
<code><a rd-options='' href='predict.glmnet'>predict.glmnet</a></code>.

predict.sentomodel: Make predictions from a sentomodel object

Description

Usage

Arguments

Value

See Also