dataConstruction.log

-------------------------------------------------------------------------------
      name:  <unnamed>
       log:  /Users/stevenchun/Google Drive/School/ECON45/Culminating/dataConst
> ruction.log
  log type:  text
 opened on:  16 May 2018, 14:15:50

. 
. 
. // Generate dummy for old variables
. clear

. import delimited using "./Dream-4-17-16.csv"
(11 vars, 30,000 obs)

. 
. duplicates drop vendor_name, force

Duplicates in terms of vendor_name

(29,032 observations deleted)

. 
. drop hash market_name item_link price name description image_link add_time sh
> ip_from v11

. 
. gen inMarket2016 = 1

. 
. save dream2016, replace
file dream2016.dta saved

. 
. // Work on 2017 cocaine prices
. clear

. 
. local shipVars ships_to_us ships_from_us ships_to_nl ///
>         ships_from_nl ships_to_fr ships_from_fr ships_to_gb ///
>         ships_from_gb ships_to_ca ships_from_ca ships_to_de ///
>         ships_from_de ships_to_au ships_from_au ships_to_eu ///
>         ships_from_eu ships_to_es ships_from_es ships_to_namerica ///
>         ships_from_namerica ships_to_be ships_from_be ships_to_ww ///
>         ships_from_ww ships_to_si ships_from_si ships_to_it ships_from_it ///
>         ships_to_dk ships_from_dk ships_to_samerica ships_from_samerica ///
>         ships_to_ch ships_from_ch ships_to_br ships_from_br ships_to_cz ///
>         ships_from_cz ships_to_se ships_from_se ships_to_co ships_from_co ///
>         ships_to_cn ships_from_cn ships_to_pl ships_from_pl ships_to_gr ///
>         ships_from_gr 

. 
. local encodedShipVars shipsToUS shipsFromUS shipsToNL ///
>         shipsFromNL shipsToFR shipsFromFR shipsToGB ///
>         shipsFromGB shipsToCA shipsFromCA shipsToDE ///
>         shipsFromDE shipsToAU shipsFromAU shipsToEU ///
>         shipsFromEU shipsToES shipsFromES shipsToNAMERICA ///
>         shipsFromNAMERICA shipsToBE shipsFromBE shipsToWW ///
>         shipsFromWW shipsToSI shipsFromSI shipsToIT shipsFromIT ///
>         shipsToDK shipsFromDK shipsToSAMERICA shipsFromSAMERICA ///
>         shipsToCH shipsFromCH shipsToBR shipsFromBR shipsToCZ ///
>         shipsFromCZ shipsToSE shipsFromSE shipsToCO shipsFromCO ///
>         shipsToCN shipsFromCN shipsToPL shipsFromPL shipsToGR ///
>         shipsFromGR

. 
. local n : word count `shipVars'

. 
. import delimited using "./data/dream_market_cocaine_listings.csv" 
(64 vars, 1,504 obs)

. 
. forvalues i = 1/`n' {
  2.     local a : word `i' of `shipVars'
  3.     local b : word `i' of `encodedShipVars'
  4.         di "`a' `b'"
  5.         gen `b' = (`a' == "TRUE")
  6.         //encode `a', generate(`b')
.         drop `a'
  7. }
ships_to_us shipsToUS
ships_from_us shipsFromUS
ships_to_nl shipsToNL
ships_from_nl shipsFromNL
ships_to_fr shipsToFR
ships_from_fr shipsFromFR
ships_to_gb shipsToGB
ships_from_gb shipsFromGB
ships_to_ca shipsToCA
ships_from_ca shipsFromCA
ships_to_de shipsToDE
ships_from_de shipsFromDE
ships_to_au shipsToAU
ships_from_au shipsFromAU
ships_to_eu shipsToEU
ships_from_eu shipsFromEU
ships_to_es shipsToES
ships_from_es shipsFromES
ships_to_namerica shipsToNAMERICA
ships_from_namerica shipsFromNAMERICA
ships_to_be shipsToBE
ships_from_be shipsFromBE
ships_to_ww shipsToWW
ships_from_ww shipsFromWW
ships_to_si shipsToSI
ships_from_si shipsFromSI
ships_to_it shipsToIT
ships_from_it shipsFromIT
ships_to_dk shipsToDK
ships_from_dk shipsFromDK
ships_to_samerica shipsToSAMERICA
ships_from_samerica shipsFromSAMERICA
ships_to_ch shipsToCH
ships_from_ch shipsFromCH
ships_to_br shipsToBR
ships_from_br shipsFromBR
ships_to_cz shipsToCZ
ships_from_cz shipsFromCZ
ships_to_se shipsToSE
ships_from_se shipsFromSE
ships_to_co shipsToCO
ships_from_co shipsFromCO
ships_to_cn shipsToCN
ships_from_cn shipsFromCN
ships_to_pl shipsToPL
ships_from_pl shipsFromPL
ships_to_gr shipsToGR
ships_from_gr shipsFromGR

. 
. //Fix prices!!
. //                                                                           
>                                    1.5 then 1,5 then .5 then ,5 then 5
. gen stringGrams = regexs(0) if(regexm(product_title, "[0-9]\.[0-9]+[ ]?[Gg]|[
> 0-9]\,[0-9]+[ ]?[Gg]|\.[0-9]+[ ]?[Gg]|\,[0-9]+[ ]?[Gg]|\.?[0-9]+[ ]?[Gg]"))
(29 missing values generated)

. 
. gen stringKilos = regexs(0) if(regexm(product_title, "[0-9]\.[0-9]+[ ]?[Kk]|[
> 0-9]\,[0-9]+[ ]?[Kk]|\.[0-9]+[ ]?[Kk]|\,[0-9]+[ ]?[Kk]|\.?[0-9]+[ ]?[Kk]"))
(1,473 missing values generated)

. 
. // destring periods and commas
. destring stringGrams, ignore(" " "g" "G") generate(numGrams) force
stringGrams: contains nonnumeric characters not specified in ignore()
stringGrams: characters space g G removed; numGrams generated as double
(74 missing values generated)

. 
. destring stringGrams, ignore(" " "g" "G") dpcomma generate(numCommaGrams) for
> ce
stringGrams: contains nonnumeric characters not specified in ignore()
stringGrams: characters space g G removed; numCommaGrams generated as double
(291 missing values generated)

. 
. // destring kils
. 
. destring stringKilos, ignore(" " "k" "K") generate(numKilos)
stringKilos: characters space k K removed; numKilos generated as byte
(1473 missing values generated)

. 
. // set fixed Grams to the appropriate amoun
. gen fixedGrams = numCommaGrams
(291 missing values generated)

. replace fixedGrams = numGrams if (missing(fixedGrams))
(262 real changes made)

. replace fixedGrams = numKilos * 1000 if (missing(fixedGrams))
(29 real changes made)

. 
. // recalculate btc prices
. replace grams = fixedGrams
(52 real changes made)

. replace cost_per_gram = btc_price / fixedGrams
(319 real changes made)

. replace cost_per_gram_pure = btc_price / (fixedGrams *(quality / 100))
(381 real changes made)

. 
. // From Coindex, average for the week the scraper was running.
. local btcInDollars = 2284.165

. 
. // Convert to US dollars
. gen usPrice = btc_price * `btcInDollars'

. gen usPricePerGram = cost_per_gram * `btcInDollars'

. gen usPricePerGramPure = cost_per_gram_pure * `btcInDollars'

. 
. // Generate flake dummy
. gen flake = regexm(lower(product_title), "fish|scale|flake")

. 
. // Shipping dummies
. 
. 
. // Generate logs
. gen lnUSPricetPerGram = ln(usPricePerGram)

. gen lnUSPricePerGramPure = ln(usPricePerGramPure)

. gen lnSuccessfulTrans = ln(successful_transactions)

. gen lnRating = ln(rating)

. 
. // Generate interactions
. //gen escrowlnSuccessfulTrans = lnSuccessfulTrans * escrow
. egen medianTransactions = median(successful_transactions)

. gen youngVendor = successful_transactions < medianTransactions

. gen escrowYoung = escrow*youngVendor

. gen ratingYoung = rating*youngVendor

. 
. 
. 
. table escrow, content(mean rating sd rating mean cost_per_gram sd cost_per_gr
> am mean cost_per_gram_pure)

----------------------------------------------------------------------------
   escrow | mean(rating)   sd(rating) mean(cost~m) sd(cost_p~m) mean(cost~e)
----------+-----------------------------------------------------------------
        0 |     4.892276     .0962092     .0369528     .0254982     .0418672
        1 |     4.881744     .1282268     .0313348     .0215021      .035568
----------------------------------------------------------------------------

. // sum rating age payoff in_market
. 
. //drop unneeded variables
. drop fixedGrams

. drop numKilos

. drop numCommaGrams

. drop numGrams

. drop stringGrams

. drop stringKilos

. drop medianTransactions

. 
. // Merge the datasets
. 
. merge m:1 vendor_name using dream2016.dta

    Result                           # of obs.
    -----------------------------------------
    not matched                         2,183
        from master                     1,247  (_merge==1)
        from using                        936  (_merge==2)

    matched                               257  (_merge==3)
    -----------------------------------------

. drop if _merge == 2
(936 observations deleted)

. 
. replace inMarket2016 = 0 if (inMarket2016 == .)
(1,247 real changes made)

. 
. save cocainePrices, replace
file cocainePrices.dta saved

. 
. 
. 
. 
. 
. 
. 
. 
. capture log close