Zusammenfassung der Ressource
Met-Sensitivity application - V0
- 1. Upload user economic
data
- 2. Select Weather
Data set from the list
- GHCN
- select "Weather Station"
- select out of list eigther
by station NAME
eigther by USAF
- National Index
- Reanalysis
- GSOD
- select weather
station
- select out of list eigther by
station NAME either by
srtation USAF either WBAN
- National Index
- SYNOP_MF
- National Index
- select "Weather Station"
- select out of list
eigther by station
NAME, eigther by
USAF
- DWD
- 4. Select the columns within your data set
- plot: time
evolution of the
user parameter
- 5. BUTTON: Data Recording
- 8. Select your time resolution
- monthly
- From month :
- to month
- daily
- 7. BUTTON: Temporal Aggregation
- 8. BUTTON : Period validation
???
- 9. Select: Regression Degree : 1 by
default
- If KNMI weather
data at ORLY:
ERROR message:
NO data
- GSOD ...
- 10. Select one or several Explanatory Variable :
Promotions, Public holidays, Day of week, Day
of month: for VISUALISATION highlighting not
for data removal
- 11. BUTTON: Start Regressions: USER data vs METEO data
- user DATA summary
TABLE: number of values,
missing values, MIN,
MAX, MEAN, Volatility
???
- RESULTS
- Summary
- 16. Select user
data column
- Histogram-type plot of the MAE :
comparison between 30 meteio
parameters:: GREEN is ??? Yellow
columns ???
- Histogram-type plot of R2
:comparison between client data
and 30 weather paramenets.
GREEN ??? Yellow ???
- 17. Download Results
- Details
- 12. Select variable: by default it is
the one with the best MAE
- 13. OPTIONAL : Highlight calendar
explanatory varuables
- YES
- 14 SELECT:
Public
Holidays,
Promotions
Day of year,
Day of week,
Day of month,
Week of year,
Quarter of yera,
Month of year,
years
- NO
- Scatter - PLOT &
TABLE : best meteo
variable vs user
data
- 15 Download results
- Custom Regression
- Selected variable : out of a long
meteo-variable list. Nothing
changes when chosing variables
??
- wished regression ??
- ??
- Launch custom regression
- Download results
- Risk period
beginning
- ??
- quaterly
- ??
- weekly
- ??
- 3. Select the UNITS
representative for
your data
- KNMI