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.

Reply via email to