Editing_A_list.qmd

# Editing list

## Chapter Specific Issues 

### Preface

A: improve install GUI link (don't leave raw link to GitHub): ok 

A: revamp tutorials (jd and not only sa) : ok 

A: revamp rjdemetra general readme, and specific ones 

### How to Use this Book

A: missing points, cf comments

### G-news-v3-vs-v2.qmd

- news v3 v2, set links to relevant part of new features 

### G-jd-history.qmd (not in book yet )

to be organized

resources - small publi 4p insee/eurostat 

web explo stack

### In All ALGO chapters 

-   R version 3 code to be added
-   remove theory related parts
-   much less text than current doc (too long)
-   method details: to method chapters
-   tools details in tools: GUI or R

when describing parameters

-   1 common description
-   2 setting in GUI
-   3 setting in R


### A-sa-overview

### A-sa-pre-treatment

- links to calendar correction 

- ramps 

graph and formula are available in [Linearisation with the TRAMO and RegARIMA models](../theory/SA_lin.html).

explain params 

- differentiation orders D1DS: if ticked: 1 1 automatically (pb differentiation before the real differentiation ?), cf rjd3toolkit doc to harmonize 

#### in reg arima results

-   seasonality tests on residuals (Highlight differences: - in gui - in R v2: retrieve object in output, v3 run test)

#### Allocation to components

allocation when intervention or ramps ? in test allocated to trend ? (reg)

impossible (?) to create several intervention variables

### A-sa-x11-decomposition

### A-sa-seats-decomposition

### A-rev-policies 

add - v3 visuals (progressive) 
- fixed auto reg param - inter chapter links : r packs, cruncher

obj here: general explanations + examples ? here : explain voc discrepancies vs guidelines bbk controlled current link to plug in illustration links on covid

#### Revision Policies

general concepts - identified vs estimated - linearization phase - X11 - seats

here concepts of domain spec, point spec

concurrent = first estimation or redo everything (never overrides user defined parameters)

#### Implemententing revision policies in R

add examples 

add example with rjd3workspace

### A-sa-hf.qmd

### A-outlier-detection.qmd

data type

contrasting GUI and R packages

small chunks on regressors to be added

more info on detection

### A-calendar-correction.qmd

ADD tool map Calendar correction using R packages vs GUI: quick contrasting

- is group and reference day choice as flexible in GUI as in R ?

- in GUI (v3) automatic detection, what in R ?

- are the same data frequencies available in R and in GUI (divisors of 12 and...)

- calendar correction HF data: R vs GUI ? including weekly data

remarks:

-   less dense content, less text more tables, bullet points

-   check and add documents new v3 features

-   cjo frequency explained by K they simulate calendar pattern...empirically fall here (cf refs papers)

frequency 41 et 41 survival of 1/60 discrete step in X12 (cf sokup findley)

#### To do

On theory part:

-   manque effet de paques (cf regressor and ICA: hole)

-   length of period

-   stock

On practical part:

-   HF in GUI ?

-   add in R

-   generating regressors from customized calendar : GUI and R

-   Length of Period

adjust param

-   stcok series

cf book x12 and SH

The regressor will have the value -1 if the w-th day is a Sunday, 1 if it is a Monday as 0 otherwise.

example lacking in the package

explain problems and mixing with seasonality ...

### A-benchmarking.qmd

review maria's work (cfr email remarks)

add rjd3 bench and link help pages

resources

-   corentin L , beamer rome

-   maria, ine, beamer ocde

### A-trend-cycle-estimation.qmd

see alain

see karsten paper on HF

### A-nowcasting.qmd

plug in v2
package v3

## T-graphical-user-interface.qmd

why use the graphical user interface ? what is not directly available in R yet?

objective: describe the general features (independent of algorithms)

-   general layout

-   options !: like stats tools/ modifs good bad / look and feel

-   import data

-   documents

-   workspaces

-   specifications

-   output


### GUI: SA and Modelling Features

chapter name/scope to change: encompass all stat methods

very weak 

lacks new links 

Topics voir what should be tackled here 

- Modelling process new document
<!-- very short explanations, as the same as in pre-adjustment  -->

- Anomaly detection (just structure link to outlier chapter)

- SAP detailed (but not algo specific), just structure 

- Seasonality tests 

- direct indirect seasonal adjustment 

 
### ADD : List of Output items

### T-Production-tools-cruncher-QR

add - List of exportable series (tramo / x13 with names)

-   link to option change for adjusting "good", "bad"...

### T-r-packages

-   add a global description of rjd3toolkit cf aux txt file

-   include details in rjd3workspace (when ready)

#### Perimeter definition

X11+

- will take all x11 related functions from rjd3highfreq which will stay model based (then seats+ will comme, another pack or not ?)

- simple (extended) airline processing + more choices in all asymetrical filters (better for revisions)**


### T-plug-ins

- change order v3 first

- add SDMX specifs 



### M-spectral-analysis

add: - R code or references to - rjd3sa (?) references to tests - more explanations on spectral analysis rationale

Vocabulary precision and concise definitions - spectral density - periodogram - spectral density estimator - "spectrum" is used too ubiquitously

### M-reg-arima-modelling

In the chapter on SA, in the pre-adjustment section, we tackle: - purpose, principles and results of reg-arima models (tramo or reg-arima)

Here we tackle : - more in depth presentation - details on model estimation (not your typical box and Jenkins)

### M-X11-decomposition.qmd

Questions

- is the A part really used in X-11 model

- in step B: "and, if requested, a first estimation of the
calendar effects" (from old doc: true ?)"


- final tables and effects of pre adjustement 

-   add more computation details: cf my notes on X-11, yes, but not urgent

Edit content goal of the chapter: details on X-11 which won't be in the SA chapter

see - my notes on X-11 - formulas from training yc estp - formulas from HF paper or beamer

### M-STL-decomposition

resources : - papier HF - papier Daniel O on dsa

### M-SEATS-decomposition

- see doc JP (new ?)

### M-Trend-Estimation-Local-Polynomials

see with KW / AQLT

### M-tests.qmd

cf wiki old doc: complement existing info 

### M-state-space-framework

cf wiki old doc 

### references

set up anew