-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathAUX-BBk-Plugins.qmd
63 lines (39 loc) · 1.48 KB
/
AUX-BBk-Plugins.qmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
# BBK {.unnumbered #bbk-plugins}
## [CompRes](https://github.com/bbkrd/CompRes)
(idea do this in a package for refresh in R: simpler ?)
additional remarks on CompRes use
(see if something neeeds to be added to BBK doc)
the point is to do controlled current:
- by default previously stored factors are kept
- confronted with update (= control): concurrent or last outliers
SEE daniel: review heuristics for updating
concurrent (= real concurrent or any refresh of factors)= blue ?
current= what was saved last (can date back to the last annual campaign, or the last infra-monthly review)
### Procedure
- current factors have to be saved to ws
- calendar
- seasonal (save and displayed are D10, not final S), test SO effect
- refresh data : ex last out or concurrent
blue = new
red= old
- tweak your refreshed spec
- two solutions
- keep new (save to ws)
- or keep old if un touched (what if I crunch, i'll get the new: YES, JD+ normal
I will have to use the excel adpater (cf belon) to compute SA with current seasonal factors
### Display
CHARTS
in charts SA = SA + C
only SA = only S without C
SI ratio
- points et croix ?
CCA ??
Table
- factors are always (!) multiplied by 100 if modele mult
## Excel adapter
for CompRes ?
compile
compute : SA
## Spec Parser
obj take an x13 census spec and import to JD+
in X13 world: 3 files series, spec, mta (equivalent of multi processing if you want to launch several series at the same time)