Il rapporto dettagliante-produttore nell'industria della moda: un'analisi basata su regressioni panel dinamiche

PARENT")

hist(challenging$PARENT_TOT, freq= FALSE, breaks= c(0,

0.5,1.5,2.5,3.5,4.5,5.5, 6.5),col = colors, main =

"Histogram of 'Challenging' PARENT-TOT", xlab = "TOT. CASES

WITH PARENT")

hist(others$PARENT_TOT, freq= FALSE, breaks= c(0,

0.5,1.5,2.5,3.5,4.5,5.5, 6.5),col = colors, main =

"Histogram of 'Others' PARENT-TOT", xlab = "TOT. CASES WITH

PARENT")

# histograms of the total cases for each reason

hist(quiet$REASON_PRODOTTI, freq= FALSE, breaks= c(0,

0.5,1.5,2.5,3.5,4.5),col = colors, main = "Histogram of

'Quiet' Reason_Prodotti", xlab = "Cases With Reason

107

Prodotti")

hist(challenging$REASON_PODOTTI, freq= FALSE, breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5, 16.5,17.5,18.5,19.5),col = colors, main =

"Histogram of 'Challenging' Reason_Prodotti", xlab = "Cases

With Reason Prodotti")

hist(others$REASON_PRODOTTI, freq= FALSE, breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5, 16.5,17.5,18.5,19.5),col = colors, main =

"Histogram of 'Others' Reason_Prodotti", xlab = "Cases With

Reason Prodotti")

hist(quiet$REASON_ORDINI, freq= FALSE, col = colors, main =

"Histogram of 'Quiet' Reason_Ordini", xlab = "Cases With

Reason Ordini")

hist(challenging$REASON_ORDINI, freq= FALSE, col = colors,

main = "Histogram of 'Challenging' Reason_Ordini", xlab =

"Cases With Reason Ordini")

hist(others$REASON_ORDINI, freq= FALSE,col = colors, main =

"Histogram of 'Others' Reason_Ordini", xlab = "Cases With

Reason ordini")

hist(quiet$REASON_AMM, freq= FALSE, breaks= c(0,

0.5,1.5,2.5,3.5,4.5,5.5), col = colors, main = "Histogram

of 'Quiet' Reason_Amministrazione", xlab = "Cases With

Reason Amministrazione") 108

hist(challenging$REASON_AMM, freq= FALSE, breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5)

, col = colors, main = "Histogram of 'Challenging'

Reason_Amministrazione", xlab = "Cases With Reason

Amministrazione")

hist(others$REASON_AMM, freq= FALSE,breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5, 16.5),col = colors, main = "Histogram of

'Others' Reason_Ordini", xlab = "Cases With Reason ordini")

hist(quiet$`REASON_HELP/INFO`, freq= FALSE, breaks= c(0,

0.5,1.5,2.5,3.5,4.5,5.5), col = colors, main = "Histogram

of 'Quiet' Reason_Help/Info", xlab = "Cases With Reason

Help/Info")

hist(challenging$`REASON_INFO/HELP`, freq= FALSE, breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5), col = colors, main = "Histogram of

'Challenging' Reason_Help/Info", xlab = "Cases With Reason

Help/Info")

hist(others$`REASON_HELP/INFO`, freq= FALSE,breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5, 16.5,17.5,18.5,19.5,20.5,21.5,22.5),col =

colors, main = "Histogram of 'Others' Reason_Help/Info",

xlab = "Cases With Reason Help/Info")

hist(quiet$REASON_QUALITY, freq= FALSE, breaks=

109

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5), col = colors, main = "Histogram of 'Quiet'

Reason_quality/Problem", xlab = "Cases With Reason

Quality/Problem")

hist(challenging$REASON_QUALITY, freq= FALSE, breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5,

16.5,17.5,18.5,19.5,20.5,21.5,22.5,23.5,24.5,25.5,26.5,27.5

,28.5,29.5,30.5,31.5,32.5,33.5), col = colors, main =

"Histogram of 'Challenging' Quality/Problem", xlab = "Cases

With Reason Quality/Problem")

hist(others$REASON_QUALITY, freq= FALSE,breaks=

c(0,0.5,1.5,2.5,3.5,4.5,5.5,6.5,7.5,8.5,9.5,10.5,11.5,12.5,

13.5,14.5,15.5,

16.5,17.5,18.5,19.5,20.5,21.5,22.5,23.5,24.5,25.5,26.5,27.5

,28.5,29.5),col = colors, main = "Histogram of 'Others'

Reason_Quality/Problem", xlab = "Cases With Reason

Quality/Problem")

• Code used for regression models with dependent variable Total Amount

ID_mt <- factor(totamount$ID)

Mese_mt <- factor(totamount$Mese)

dataset_fact_t <- totamount[,-1]

dataset_fact_t <- dataset_fact_t[,-1]

110

dataset_finale_t <- data.frame(ID_mt , Mese_mt ,

dataset_fact_t)

dataset_finale_t$TotAmount = dataset_finale_t$TotAmount/100

panel_mt <- pdata.frame(dataset_finale_t, index =

c("ID_mt", "Mese_mt"))

# Random effects model

RAND_m3<- plm(TotAmount ~

Status_CLOSED+Solution_SOLVED+Parent_C+Reason_PRODOTTI_C+Re

ason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True+ CASE_TOT_CREATI, data= panel_mt, model= "random")

# Fixed effects model

fix_m3<- plm(TotAmount ~

Status_CLOSED+Solution_SOLVED+Parent_C+Reason_PRODOTTI_C+Re

ason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True+ CASE_TOT_CREATI, data= panel_mt, model= "within")

# Hausman Test

phtest(fix_m3,RAND_m3)

# Fixed effects model with robust errors

coeftest(fix_m3,vcovHC) 111

# Arellano-Bond model (with different lags)

din_1_m3<- pgmm(TotAmount ~

lag(TotAmount,1)+CASE_TOT_CREATI+

Status_CLOSED+Solution_SOLVED+Solution_NOTSOLVED+Parent_C+R

eason_PRODOTTI_C+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 2:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_2_m3<- pgmm(TotAmount ~

lag(TotAmount,2)+CASE_TOT_CREATI+

Status_CLOSED+Solution_NOTSOLVED+Parent_C+Reason_PRODOTTI_C

+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 3:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_3_m3<- pgmm(TotAmount ~

lag(TotAmount,3)+CASE_TOT_CREATI+

Status_CLOSED+Solution_NOTSOLVED+Parent_C+Reason_PRODOTTI_C

+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 4:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_4_m3<- pgmm(TotAmount ~ 112

lag(TotAmount,4)+CASE_TOT_CREATI+

Status_CLOSED+Solution_NOTSOLVED+Parent_C+Reason_PRODOTTI_C

+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 5:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_5_m3<- pgmm(TotAmount ~

lag(TotAmount,1:2)+CASE_TOT_CREATI+

Status_CLOSED+Solution_SOLVED+NOSOL_Fraz+Parent_C+Reason_PR

ODOTTI_C+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 3:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_6_m3<- pgmm(TotAmount ~

lag(TotAmount,1:3)+CASE_TOT_CREATI+

Status_CLOSED+Solution_SOLVED+NOSOL_Fraz+Parent_C+Reason_PR

ODOTTI_C+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 4:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_7_m3<- pgmm(TotAmount ~

lag(TotAmount,1:4)+CASE_TOT_CREATI+

Status_CLOSED+Solution_SOLVED+NOSOL_Fraz+Parent_C+Reason_PR

113

ODOTTI_C+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 5:11), data= panel_mt, model =

"twosteps", effect = "individual")

din_8_m3<- pgmm(TotAmount ~

lag(TotAmount,1:5)+CASE_TOT_CREATI+

Status_CLOSED+Solution_SOLVED+NOSOL_Fraz+Parent_C+Reason_PR

ODOTTI_C+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(TotAmount, 6:11), data= panel_mt, model =

"twosteps", effect = "individual")

• Code used for regression models with dependent variable “Cases created”

ID_m3 <- factor(Cartel4$ID)

Mese_m3 <- factor(Cartel4$Mese)

dataset_fact_3 <- Cartel4[,-1]

dataset_fact_3 <- dataset_fact_3[,-1]

dataset_finale_m3 <- data.frame(ID_m3 , Mese_m3 ,

dataset_fact_3)

dataset_finale_m3$TotAmount =

dataset_finale_m3$TotAmount/100

panel_m3 <- pdata.frame(dataset_finale_m3, index =

c("ID_m3", "Mese_m3")) 114

# Poisson model with random effects

pois_random_m3<- pglm(CASE_TOT_CREATI ~

Status_CLOSED+Solution_SOLVED+Parent_C+Reason_PRODOTTI_C+Re

ason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True+TotAmount, data= panel_m3, model= "random", family

= "poisson")

# Poisson model with fixed effects

pois_fixed_m3<- pglm(CASE_TOT_CREATI ~

Status_CLOSED+Solution_SOLVED+Parent_C+Reason_PRODOTTI_C+Re

ason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True+TotAmount, data= panel_m3, model= "within", family

= "poisson")

# Hausman Test

phtest(pois_fixed_m3,pois_rand_m3)

# Fixed effects model with robust errors

coeftest(pois_fixed_m3, vcov = sandwich)

# Arellano-Bond model (with different lags)

din_1_m3<- pgmm(CASE_TOT_CREATI ~

lag(CASE_TOT_CREATI,1)+TotAmount+

115

Status_CLOSED+Solution_SOLVED+Solution_NOTSOLVED+Parent_C+R

eason_PRODOTTI_C+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(CASE_TOT_CREATI, 2:11), data= panel_m3, model =

"twosteps", effect = "individual")

din_2_m3<- pgmm(CASE_TOT_CREATI ~

lag(CASE_TOT_CREATI,2)+TotAmount+

Status_CLOSED+Solution_NOTSOLVED+Parent_C+Reason_PRODOTTI_C

+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(CASE_TOT_CREATI, 3:11), data= panel_m3, model =

"twosteps", effect = "individual")

din_3_m3<- pgmm(CASE_TOT_CREATI ~

lag(CASE_TOT_CREATI,3)+TotAmount+

Status_CLOSED+Solution_NOTSOLVED+Parent_C+Reason_PRODOTTI_C

+Reason_ORDINI_C+Reason_AMMINISTRAZIONE_C +

Reason_HELP.INFO_C+`Reason_QUALI.PROBL_C`+SLA_Resp+OneConta

ct_True|lag(CASE_TOT_CREATI, 4:11), data= panel_m3, model =

"twosteps", effect = "individual")

din_4_m3<- pgmm(CASE_TOT_CREATI ~

lag(CASE_TOT_CREATI,4)+TotAmount+

Status_CLOSED+Solution_NOTSOLVED+Parent_C+Reason_PRODOTTI_C

+Reason_