Open robj411 opened 5 years ago
I think trip data is sorted (see here: https://github.com/ITHIM/ITHIM-R/commit/5520275c8532b4b6ba16c533494fe26408c54628)
@robj411 just re-reading the list. Found that Bangalore population dataset
was not an outstanding issue. Don't think we used some place else's population for it.
updated results on 25/04
Looking at the distance (km) and duration (mins) for short walk
assosicated with bus
trips, is as follows:
(units: mins)
summary(rd %>% filter(trip_mode == "bus" & stage_mode == "short walk") %>% select(stage_duration)) stage_duration Min. : 2.00
1st Qu.: 4.00
Median :10.00
Mean :10.55
3rd Qu.:15.00
Max. :70.00
(units: kms)
summary(rd %>% filter(trip_mode == "bus" & stage_mode == "short walk") %>% select(stage_distance)) stage_distance
Min. :0.1600
1st Qu.:0.3200
Median :0.8000
Mean :0.8439
3rd Qu.:1.2000
Max. :5.6000
In terms of distribution, this is how the density plots look like:
@leandromtg SP's trip data is located at: inst/extdata/local/sao_paulo/trips_sao_paulo.csv
Ali, can you confirm if these numbers should be per trip or person? Durations are quite high anyway and don't match distance stats. I will do for Sao Paulo early next week.
@leandromtg thanks for looking at the distribution. You were right - they were incorrect. Now have fixed them. Please see the plots above.
Re these numbers being per trip or per person, it is per trips. I believe this is what we need, don't we?
Ali, can you confirm if these numbers should be per trip or person?
Durations are quite high anyway and don't match distance stats.
I will do for Sao Paulo early next week.
Thanks,
Leandro
On Thu, 18 Apr 2019, 15:29 Ali Abbas, notifications@github.com wrote:
Looking at the distance (km) and duration (mins) for short walk assosicated with bus trips, is as follows:
stage_duration Min. : 5.00 1st Qu.: 30.00 Median : 50.00 Mean : 58.62 3rd Qu.: 80.00 Max. :240.00
stage_distance Min. :0.1600 1st Qu.:0.3200 Median :0.8000 Mean :0.8439 3rd Qu.:1.2000 Max. :5.6000
In terms of distribution, this how the density plots look like:
[image: image] https://user-images.githubusercontent.com/10958628/56367977-6704f680-61ee-11e9-98e6-d336078c9739.png
[image: image] https://user-images.githubusercontent.com/10958628/56368022-81d76b00-61ee-11e9-9fdc-c9c341b0d3fd.png
@leandromtg https://github.com/leandromtg SP's trip data is located at: inst/extdata/local/sao_paulo/trips_sao_paulo.csv
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/ITHIM/ITHIM-R/issues/55#issuecomment-484534707, or mute the thread https://github.com/notifications/unsubscribe-auth/ACTUQBZS567HBKMJAQGRLUDPRCAVLANCNFSM4HBVWIYQ .
This is a to-do list for the Bangalore setting, which includes data, parameters, and any special functions required. Please add to it any items that are missing, and tick off items when they are complete.
[ ] PA data (pa_bangalore.csv file with columns: "sex", "age", "work_ltpa_travel_marg_met") Using Sao Paulo data for now, @rahulatiitd and @leandromtg to replace this with India data (SAGE 2007)
[x] Get GBD data for Bangalore, rather than India (gbd_bangalore.csv) OR scale/adjust India data @rahulatiitd to contact GBD team to get detailed data for Delhi (age and sex specific)
[x] Bangalore injury dataset
[ ] Bangalore trips dataset
[x] Bangalore population dataset
[ ] distribution for BUS_WALK_TIME, if walk to bus is added @usr110 and @leandromtg to look at this distribution from reported numbers in Sao Paulo @rahulatiitd also to estimate this for Delhi
[x] distribution for PM_CONC_BASE (background PM2.5) mean 47.7 distribution lnorm(log(47.4),log(1.17))
[x] distribution for PM_TRANS_SHARE (transport share of PM2.5) mean 0.281 distribution beta(7,18)
[ ] distribution for BACKGROUND_PA_SCALAR (amount to scale non-zero, non-travel PA)
[ ] number for BACKGROUND_PA_CONFIDENCE (how confident are you about the number of zeros in the PA survey on a scale of 0 to 1?)
[x] distribution for INJURY_REPORTING_RATE (rate at which we think injuries were reported) beta(50,3)
[ ] distribution for CHRONIC_DISEASE_SCALAR (amount to scale NCDs)
[x] distribution for DAY_TO_WEEK_TRAVEL_SCALAR 7
[x] distribution for BUS_TO_PASSENGER_RATIO (if bus drivers are added) same as sp -- beta(20,600), 98.5% between 1/20 and 1/60
[x] distribution for TRUCK_TO_CAR_RATIO (if truck drivers are added). same as sp -- Beta(3,10)