On the Estimation of Security Price Volatilities from Historical Data

TLDR

The study focuses on volatility estimation using data that are universally accessible to investors, namely the high, low, opening, and closing prices and transaction volume reported in newspaper financial pages, while excluding other data sources such as news events or fundamental information. It aims to estimate capital asset price volatility parameters by developing and applying estimators based on these publicly available OHLC and volume figures. The authors formulate improved volatility estimators within a structural model detailed in Section II, and present their derivation and implementation in Section III using the aforementioned newspaper data. These new estimators exhibit considerably higher relative efficiencies than standard volatility estimators.

Abstract

This paper examines the problem of estimating capital asset price volatility parameters from the most available forms of public data. While many varieties of such data are possible, we shall consider here only those which are truly universal in their accessibility to investors, namely, data appearing in the financial pages of the newspaper. In particular, we shall consider volatility estimators which are based upon the historical opening, closing, high, and low prices and transaction volume. Alternative estimators of volatility may be constructed from such data as significant news events, fundamental information regarding a company's prospects, and other forms of publicly available data, but these will not be considered here. Any parameter-estimation procedure must begin with a maintained hypothesis regarding the structural model within which estimation is to be made. Our structural model is given exposition in Section II. Section III discusses the classical Improved estimators of security price volatilities are formulated. These estimators employ data of the type commonly found in the financial pages of a newspaper: the high, low, opening, and closing prices and the transaction volume. The new estimators are seen to have relative efficiencies that are considerably higher than the standard estimators.