A question about vectorized operations (avoiding loops, for speed)... I need to run a simple recursive (autoregressive) filter with a time-varying coefficient. It is just a one-step recursive filter, so it would be an exponential decay if the filter was constant. I just want to do this, where 'x' is the data and 'w' is the weight to apply to the previous time step: x <- c(1, 1, 0, 2, 0, 0) w <- c(NA, 0.1, 0.5, 0.4, 0.3, 0.2) y <- x for (i in seq_along(x)[-1]) y[i] <- y[i] + w[i] * y[i-1] print(y) [1] 1.0000 1.1000 0.5500 2.2200 0.6660 0.1332 But, since loops are slow, I would like a vectorized method, like filter(, method="recursive"). Any ideas?
Felix Andrews / ??? PhD candidate, The Fenner School of Environment and Society The Australian National University (Building 48A), ACT 0200 Beijing Bag, Locked Bag 40, Kingston ACT 2604 http://www.neurofractal.org/felix/ voice:+86_1051404394 (in China) mobile:+86_13522529265 (in China) mobile:+61_410400963 (in Australia) xmpp:foolish.android at gmail.com 3358 543D AAC6 22C2 D336 80D9 360B 72DD 3E4C F5D8