Hi, Does anyone know how to compute "price velocity" and "price acceleration" for a financial time series, or what package implements this (maybe under another name)?
I am trying to apply a linear regression from the following text where both terms are coefficients in the regression: "The 18 candidate price dynamics indicators are of 3 types: velocity, acceleration and volatility, with six variants of each type. The variants differ with respect to the number of days used to measure velocity and acceleration or with respect to the exponential smoothing constant used to measure volatility. Type 1 (*price velocity*) is the slope term of a moving linear regression, fit using least squares, to the logs of the S&P500 close. The six fitting or look-back periods are 11, 22, 44, 65, 130 and 260 days. Specifically, we define price velocity as the coefficient b in the function y =a +bx, where y is the log of price and x is the date index (increasing by one for each trade date). Type 2 (*price acceleration or curvature*) is the second order term of a moving parabolic regression, fit using least squares to the logs of the S&P500 close using fitting periods of 11, 22, 44, 65, 130 and 260 days. Thus acceleration is the c coefficient in the function y= a + bx + cx2 where y is the log of price and x is the date index." from: Purified Sentiment Indicators For The Stock Market http://www.hoodriverresearch.com/PurifiedSentimentIndicatorsfortheStockMarket5.04.09.pdf Regards, -Mark- [[alternative HTML version deleted]]
_______________________________________________ [email protected] mailing list https://stat.ethz.ch/mailman/listinfo/r-sig-finance -- Subscriber-posting only. If you want to post, subscribe first. -- Also note that this is not the r-help list where general R questions should go.
