ewma

Hi all, Refer to the attached file which covers EWMA, I dont get the paragraph highlighted in blue. If new weight (alpha sub i+1) = lambda x current weights (alpha sub i) where lambda is between 0 and 1, it means that new weights has a lesser weights compared to old weights. For eg, old weight...

Hi, I have a question about the EWMA model equation. How comes we are using the previous day's return? I thought we are using today's return as \[ r_n=\ln(s_n/s_{n-1}) \].

The general form for all three is: σ^2(n) = γ*V(L) + α*u^2(n-1) + σ^2(n-1).

The exponentially weighted moving average (EWMA) cures the key weakness of the common historical standard deviation by assigning greater weight to more recent returns and lessor weights to more distant (in the past) returns. Its key parameter is lambda, λ, which specifies the ratio of...

Hi, I am reviewing the EWMA model section and found returns were calculated on straight (Pt+1/Pt)-1, shouldn't the returns be calculated on log basis? Also what's the assumption for the test per se?

The 5th percentile should be between lowest and 2nd lowest transaction i.e. -4.70 % -4.10% then how -3.6% and -3.4% choosen?? Please clarify

How can I understand the notations better?