Function to set parameters by either using the constant value or sampling from a pre-defined function
Usage
ithim_setup_parameters(
NSAMPLES = 1,
BUS_WALK_TIME = 16,
RAIL_WALK_TIME = 12.5,
CYCLING_MET = 6.8,
WALKING_MET = 3.5,
PASSENGER_MET = 1.3,
CAR_DRIVER_MET = 2.5,
MOTORCYCLIST_MET = 2.8,
SEDENTARY_ACTIVITY_MET = 1.3,
LIGHT_ACTIVITY_MET = 1.3,
MODERATE_PA_MET = 4,
VIGOROUS_PA_MET = 8,
PM_CONC_BASE = 12.69,
PM_TRANS_SHARE = 0.42,
PA_DOSE_RESPONSE_QUANTILE = F,
AP_DOSE_RESPONSE_QUANTILE = F,
BACKGROUND_PA_SCALAR = 1,
BACKGROUND_PA_CONFIDENCE = 1,
INJURY_REPORTING_RATE = 1,
CHRONIC_DISEASE_SCALAR = 1,
DAY_TO_WEEK_TRAVEL_SCALAR = 7,
SIN_EXPONENT_SUM = 2,
CASUALTY_EXPONENT_FRACTION = 0.5,
SIN_EXPONENT_SUM_NOV = 1,
SIN_EXPONENT_SUM_CYCLE = 2,
CASUALTY_EXPONENT_FRACTION_CYCLE = 0.5,
SIN_EXPONENT_SUM_PED = 2,
CASUALTY_EXPONENT_FRACTION_PED = 0.5,
SIN_EXPONENT_SUM_VEH = 2,
CASUALTY_EXPONENT_FRACTION_VEH = 0.5,
BUS_TO_PASSENGER_RATIO = 0.0389,
CAR_OCCUPANCY_RATIO = 0.625,
TRUCK_TO_CAR_RATIO = 0.3,
FLEET_TO_MOTORCYCLE_RATIO = 0.441,
PROPORTION_MOTORCYCLE_TRIPS = 0,
PM_EMISSION_INVENTORY_CONFIDENCE = 1,
CO2_EMISSION_INVENTORY_CONFIDENCE = 1,
DISTANCE_SCALAR_CAR_TAXI = 1,
DISTANCE_SCALAR_WALKING = 1,
DISTANCE_SCALAR_PT = 1,
DISTANCE_SCALAR_CYCLING = 1,
DISTANCE_SCALAR_MOTORCYCLE = 1,
BUS_DRIVER_PROP_MALE = 0.98,
BUS_DRIVER_MALE_AGERANGE = "19, 65",
BUS_DRIVER_FEMALE_AGERANGE = "19, 65",
TRUCK_DRIVER_PROP_MALE = 0.98,
TRUCK_DRIVER_MALE_AGERANGE = "18, 65",
TRUCK_DRIVER_FEMALE_AGERANGE = "18, 65",
COMMERCIAL_MBIKE_PROP_MALE = 0.99,
COMMERCIAL_MBIKE_MALE_AGERANGE = "18, 65",
COMMERCIAL_MBIKE_FEMALE_AGERANGE = "18, 65",
MINIMUM_PT_TIME = 3,
MODERATE_PA_CONTRIBUTION = 0.5,
CALL_INDIVIDUAL_SIN = F,
SCENARIO_NAME = "GLOBAL",
SCENARIO_INCREASE = 0.05
)Arguments
- NSAMPLES
constant integer: number of samples to take
- BUS_WALK_TIME
lognormal parameter: duration of walk to bus stage
- RAIL_WALK_TIME
lognormal parameter: duration of walk to rail stage
- CYCLING_MET
lognormal parameter: METs when cycling
- WALKING_MET
lognormal parameter: METs when walking
- PASSENGER_MET
lognormal parameter: MET value associated with being a passenger on public transport
- CAR_DRIVER_MET
lognormal parameter: MET value associated with being a car driver
- MOTORCYCLIST_MET
lognormal parameter: MET value associated with being a motorcyclist
- SEDENTARY_ACTIVITY_MET
lognormal parameter: MET value associated with sedentary activity
- LIGHT_ACTIVITY_MET
lognormal parameter: MET value associated with light activity
- MODERATE_PA_MET
lognormal parameter: MET value associated with moderate activity
- VIGOROUS_PA_MET
lognormal parameter: MET value associated with vigorous activity
- PM_CONC_BASE
lognormal parameter: background PM2.5 concentration
- PM_TRANS_SHARE
beta parameter: fraction of background PM2.5 attributable to transport
- PA_DOSE_RESPONSE_QUANTILE
logic: whether or not to sample from physical activity relative risk dose response functions
- AP_DOSE_RESPONSE_QUANTILE
logic: whether or not to sample from air pollution relative risk dose response functions
- BACKGROUND_PA_SCALAR
lognormal parameter: reporting scalar for physical activity to correct bias in data
- BACKGROUND_PA_CONFIDENCE
beta parameter: confidence in accuracy of zero non-travel physical activity levels
- INJURY_REPORTING_RATE
lognormal parameter: rate of injury fatality reporting
- CHRONIC_DISEASE_SCALAR
lognormal parameter: scalar for background disease rates to adjust for bias in GBD data
- DAY_TO_WEEK_TRAVEL_SCALAR
beta parameter: rate of scaling travel from one day to one week - CURRENTLY used as constant only (using as beta parameter would need some further considerations)
- SIN_EXPONENT_SUM
lognormal parameter: linearity of injuries with respect to two modes. SIN_EXPONENT_SUM=2 means no safety in numbers
- CASUALTY_EXPONENT_FRACTION
beta parameter: casualty exponent contribution to SIN_EXPONENT_SUM
- SIN_EXPONENT_SUM_NOV
lognormal parameter: linearity of injuries with respect to two modes where strike mode = NOV. SIN_EXPONENT_SUM=2 means no safety in numbers
- SIN_EXPONENT_SUM_CYCLE
lognormal parameter: linearity of injuries with respect to two modes where victim mode = cycle. SIN_EXPONENT_SUM=2 means no safety in numbers
- CASUALTY_EXPONENT_FRACTION_CYCLE
beta parameter: casualty exponent contribution to SIN_EXPONENT_SUM_CYCLE where victim mode = cycle
- SIN_EXPONENT_SUM_PED
lognormal parameter: linearity of injuries with respect to two modes where victim mode = pedestrian. SIN_EXPONENT_SUM=2 means no safety in numbers
- CASUALTY_EXPONENT_FRACTION_PED
beta parameter: casualty exponent contribution to SIN_EXPONENT_SUM_PED where victim mode = pedestrian
- SIN_EXPONENT_SUM_VEH
lognormal parameter: linearity of injuries with respect to two modes where victim mode = a vehicle. SIN_EXPONENT_SUM=2 means no safety in numbers
- CASUALTY_EXPONENT_FRACTION_VEH
beta parameter: casualty exponent contribution to SIN_EXPONENT_SUM_VEH where victim mode = a vehicle
- BUS_TO_PASSENGER_RATIO
beta parameter: number of buses per passenger
- CAR_OCCUPANCY_RATIO
beta parameter: number of people per car (including driver)
- TRUCK_TO_CAR_RATIO
beta parameter: proportion of truck to car vehicle km travelled
- FLEET_TO_MOTORCYCLE_RATIO
beta parameter: amount of motorcycle trips that are to be added as commercial trips
- PROPORTION_MOTORCYCLE_TRIPS
beta parameter: proportion of trips that are to be added as personal motorcycle trips
- PM_EMISSION_INVENTORY_CONFIDENCE
beta parameter: confidence in accuracy of PM emission inventory
- CO2_EMISSION_INVENTORY_CONFIDENCE
beta parameter: confidence in accuracy of CO2 emission inventory
- DISTANCE_SCALAR_CAR_TAXI
lognormal parameter: scalar to adjust for bias in car distance travelled
- DISTANCE_SCALAR_WALKING
lognormal parameter: scalar to adjust for bias in walking distance travelled
- DISTANCE_SCALAR_PT
lognormal parameter: scalar to adjust for bias in PT distance travelled
- DISTANCE_SCALAR_CYCLING
lognormal parameter: scalar to adjust for bias in cycling distance travelled
- DISTANCE_SCALAR_MOTORCYCLE
lognormal parameter: scalar to adjust for biase in motorcycle distance travelled
- BUS_DRIVER_PROP_MALE
scalar: proportion of bus drivers that are male
- BUS_DRIVER_MALE_AGERANGE
character: age range of male bus drivers
- BUS_DRIVER_FEMALE_AGERANGE
character: age range of female bus drivers
- TRUCK_DRIVER_PROP_MALE
scalar: proportion of truck drivers that are male
- TRUCK_DRIVER_MALE_AGERANGE
character: age range of male truck drivers
- TRUCK_DRIVER_FEMALE_AGERANGE
character: age range of female truck drivers
- COMMERCIAL_MBIKE_PROP_MALE
scalar: proportion of commercial motorcycle drivers that are male
- COMMERCIAL_MBIKE_MALE_AGERANGE
character: age range of male commercial motorcycle drivers
- COMMERCIAL_MBIKE_FEMALE_AGERANGE
character: age range of female commercial motorcycle drivers
- MINIMUM_PT_TIME
scalar: minimum time that person spends on public transport
- MODERATE_PA_CONTRIBUTION
scalar: proportion contribution of moderate PA in Leisure MVPA
- CALL_INDIVIDUAL_SIN
logic: whether or not to call the safety in number coefficients for individual vehicles or use the same coefficients for all modes
Details
For each input parameters there are two options: to be set to a constant, or to be sampled from a specified distribution. Each parameter is given as an argument of length 1 or 2. If length 1, it's constant, and set to the global environment. If length 2, a distribution is defined and sampled from NSAMPLE times. There are some exceptions, listed below.
The function performs the following steps:
set all input parameters to the global environment (if sampling function is called, they are overwritten)
loop through all potential variables with a lognormal distribution and sample from this distribution if required
loop through all potential variables with a beta distribution and sample from this distribution if required
if BACKGROUND_PA_CONFIDENCE<1 then add BACKGROUND_PA_ZEROS parameters
if PM_EMISSION_INVENTORY_CONFIDENCE<1, then sample those PM inventory values by using a Dirichlet distribution which is parameterised by gamma random variables
if CO2_EMISSION_INVENTORY_CONFIDENCE<1, then sample those CO2 inventory values by using a Dirichlet distribution which is parameterised by gamma random variables
if PA_DOSE_RESPONSE_QUANTILE == T, find all diseases that are related to physical activity levels and assign a quantile to them by sampling from a uniform distribution between 0 and 1
if AP_DOSE_RESPONSE_QUANTILE == T, find all diseases that are related to air pollution levels and assign a quantile to them by sampling from a uniform distribution between 0 and 1
At the bottom of this function, the dirichlet_pointiness() function is defined which
parameterises the Dirichlet distributions for the PM and CO2 emission inventories.