Business Applications and State-Level Stock Market Realized Volatility: A Forecasting Experiment

Matteo Bonato; Oguzhan Cepni; Rangan Gupta; Christian Pierdzioch

Business Applications and State-Level Stock Market Realized Volatility: A Forecasting Experiment

Matteo Bonato, Oguzhan Cepni, Rangan Gupta, Christian Pierdzioch

Department of Economics

Research output: Working paper › Research

Abstract

We analyze the predictive value of state-level business applications, as a proxy of local investor sentiment, for the state-level realized US stock-market volatility. We use highfrequency data for the period from September, 2011 to October, 2021 to compute realized volatility. We show, using an extended version of the popular heterogenous autoregressive realized volatility model, that business applications have predictive value at intermediate and long prediction horizons, after controlling for realized moments (realized skewness, realized kurtosis, realized tail risks), for realized state-level stock-market volatility, and for upside ("good") and downside ("bad") realized volatility.

Original language	English
Place of Publication	Pretoria
Publisher	University of Pretoria
Number of pages	22
Publication status	Published - Oct 2022

Series	Working Paper Series / Department of Economics. University of Pretoria
Number	2022-47

Keywords

State-level stock markets
State-level investor sentiment
Business applications
Realized volatility
Forecasting

Access to Document

https://www.up.ac.za/media/shared/61/WP/wp_2022_47.zp225794.pdfLicence: Unspecified

Handle.net

Persistent link

Cite this

@techreport{29cfab10c8064dc58e93fa53dcc6a057,

title = "Business Applications and State-Level Stock Market Realized Volatility: A Forecasting Experiment",

abstract = "We analyze the predictive value of state-level business applications, as a proxy of local investor sentiment, for the state-level realized US stock-market volatility. We use highfrequency data for the period from September, 2011 to October, 2021 to compute realized volatility. We show, using an extended version of the popular heterogenous autoregressive realized volatility model, that business applications have predictive value at intermediate and long prediction horizons, after controlling for realized moments (realized skewness, realized kurtosis, realized tail risks), for realized state-level stock-market volatility, and for upside ({"}good{"}) and downside ({"}bad{"}) realized volatility.",

keywords = "State-level stock markets, State-level investor sentiment, Business applications, Realized volatility, Forecasting, State-level stock markets, State-level investor sentiment, Business applications, Realized volatility, Forecasting",

author = "Matteo Bonato and Oguzhan Cepni and Rangan Gupta and Christian Pierdzioch",

year = "2022",

month = oct,

language = "English",

series = "Working Paper Series / Department of Economics. University of Pretoria ",

publisher = "University of Pretoria",

number = "2022-47",

address = "South Africa",

type = "WorkingPaper",

institution = "University of Pretoria",

}

TY - UNPB

T1 - Business Applications and State-Level Stock Market Realized Volatility

T2 - A Forecasting Experiment

AU - Bonato, Matteo

AU - Cepni, Oguzhan

AU - Gupta, Rangan

AU - Pierdzioch, Christian

PY - 2022/10

Y1 - 2022/10

N2 - We analyze the predictive value of state-level business applications, as a proxy of local investor sentiment, for the state-level realized US stock-market volatility. We use highfrequency data for the period from September, 2011 to October, 2021 to compute realized volatility. We show, using an extended version of the popular heterogenous autoregressive realized volatility model, that business applications have predictive value at intermediate and long prediction horizons, after controlling for realized moments (realized skewness, realized kurtosis, realized tail risks), for realized state-level stock-market volatility, and for upside ("good") and downside ("bad") realized volatility.

AB - We analyze the predictive value of state-level business applications, as a proxy of local investor sentiment, for the state-level realized US stock-market volatility. We use highfrequency data for the period from September, 2011 to October, 2021 to compute realized volatility. We show, using an extended version of the popular heterogenous autoregressive realized volatility model, that business applications have predictive value at intermediate and long prediction horizons, after controlling for realized moments (realized skewness, realized kurtosis, realized tail risks), for realized state-level stock-market volatility, and for upside ("good") and downside ("bad") realized volatility.

KW - State-level stock markets

KW - State-level investor sentiment

KW - Business applications

KW - Realized volatility

KW - Forecasting

KW - State-level stock markets

KW - State-level investor sentiment

KW - Business applications

KW - Realized volatility

KW - Forecasting

M3 - Working paper

T3 - Working Paper Series / Department of Economics. University of Pretoria

BT - Business Applications and State-Level Stock Market Realized Volatility

PB - University of Pretoria

CY - Pretoria

ER -