1LinPred(3) User Contributed Perl Documentation LinPred(3)
2
3
4
6 PDL::Filter::LinPred - Linear predictive filtering
7
9 $a = new PDL::Filter::LinPred(
10 {NLags => 10,
11 LagInterval => 2,
12 LagsBehind => 2,
13 Data => $dat});
14
15 ($pd,$corrslic) = $a->predict($dat);
16
18 A filter by doing linear prediction: tries to predict the next value in
19 a data stream as accurately as possible. The filtered data is the pre‐
20 dicted value. The parameters are
21
22 NLags Number of time lags used for prediction
23
24 LagInterval
25 How many points each lag should be
26
27 LagsBehind
28 If, for some strange reason, you wish to predict not the next
29 but the one after that (i.e. usually f(t) is predicted from
30 f(t-1) and f(t-2) etc., but with LagsBehind => 2, f(t) is pre‐
31 dicted from f(t-2) and f(t-3)).
32
33 Data The input data, which may contain other dimensions past the
34 first (time). The extraneous dimensions are assumed to repre‐
35 sent epochs so the data is just concatenated.
36
37 AutoCovar
38 As an alternative to Data, you can just give the temporal auto‐
39 correlation function.
40
41 Smooth Don't do prediction or filtering but smoothing.
42
43 The method predict gives a prediction for some data plus a correspond‐
44 ing slice of the data, if evaluated in list context. This slice is
45 given so that you may, if you wish, easily plot them atop each other.
46
47 The rest of the documentation is under lazy evaluation.
48
50 Copyright (C) Tuomas J. Lukka 1997. All rights reserved. There is no
51 warranty. You are allowed to redistribute this software / documentation
52 under certain conditions. For details, see the file COPYING in the PDL
53 distribution. If this file is separated from the PDL distribution, the
54 copyright notice should be included in the file.
55
56
57
58perl v5.8.8 2000-05-23 LinPred(3)