ethnicity more controls

#2021 godina sa genderom i dobnim grupama
library (readxl)
X2021_medjuopcinske_Mig <- read_excel("C:/Users/Amra/OneDrive - Direkcija za ekonomsko planiranje/Prijava teme doktorske disertacije/data collection for BiH/!FBiHmedjuopstinskeunutrasnjemigracije/2021_medjuopcinske_Mig.xlsx")
X_2021 <- X2021_medjuopcinske_Mig

library(tidyr)
library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

long_2021 <- X_2021 %>% 
pivot_longer(
    cols = "BIHAĆ":"INO", 
    names_to = "iz/u",
    values_to = "value",
    names_repair = "minimal"  #veoma bitno to je kad imamo veoma sitnih razlika i onda kazemo R-u da ih zanemari
  )

# dodaj 2021 godinu
long_2021$Y <- 2021
dim(long_2021)

## [1] 454608      7

long_2021 <- long_2021 %>% 
  filter(Nivo==3) %>% 
  filter(Pol != "Ukupno") %>%
  filter (STAROST != "UKUPNO")
dim (long_2021)

## [1] 259120      7

# transform it to individual level
# Load necessary library
library(tidyr)

# Transform the dataset by expanding rows based on the 'value' column
long_2021_indiv <- long_2021 %>%
  uncount(value)

# Print the transformed dataset
print(long_2021_indiv)

## # A tibble: 21,434 × 6
##     Nivo `KANTON/OPĆINA` STAROST Pol   `iz/u`               Y
##    <dbl> <chr>           <chr>   <chr> <chr>            <dbl>
##  1     3 BOSANSKA KRUPA  0-4     M     BIHAĆ             2021
##  2     3 BOSANSKA KRUPA  0-4     M     BUŽIM             2021
##  3     3 BOSANSKA KRUPA  0-4     M     KALESIJA          2021
##  4     3 BOSANSKA KRUPA  0-4     M     TUZLA             2021
##  5     3 BOSANSKA KRUPA  0-4     M     INO               2021
##  6     3 BOSANSKA KRUPA  0-4     Ž     CAZIN             2021
##  7     3 BOSANSKA KRUPA  0-4     Ž     CAZIN             2021
##  8     3 BOSANSKA KRUPA  0-4     Ž     KLJUČ             2021
##  9     3 BOSANSKA KRUPA  0-4     Ž     REPUBLIKA SRPSKA  2021
## 10     3 BOSANSKA KRUPA  5-9     M     BIHAĆ             2021
## # ℹ 21,424 more rows

dim (long_2021_indiv)

## [1] 21434     6

#rename
long_2021_indiv <- long_2021_indiv %>% rename(from =`KANTON/OPĆINA`, to = `iz/u`)
names(long_2021_indiv)

## [1] "Nivo"    "from"    "STAROST" "Pol"     "to"      "Y"

Ponovi za 2022 godinu

#2022 godina sa genderom i dobnim grupama

library(readxl)
X2022_medjuopcinske_Mig <- read_excel("C:/Users/Amra/OneDrive - Direkcija za ekonomsko planiranje/Prijava teme doktorske disertacije/data collection for BiH/!FBiHmedjuopstinskeunutrasnjemigracije/2022_medjuopcinske_Mig.xlsx", 
    sheet = "Za R")

## New names:
## • `UKUPNO` -> `UKUPNO...56`
## • `UKUPNO` -> `UKUPNO...66`

X_2022 <- X2022_medjuopcinske_Mig

library(tidyr)
library(dplyr)
long_2022 <- X_2022 %>% 
pivot_longer(
    cols = "BIHAĆ":"INO", 
    names_to = "iz/u",
    values_to = "value",
    names_repair = "minimal"  #veoma bitno to je kad imamo veoma sitnih razlika i onda kazemo R-u da ih zanemari
  )

# dodaj 2022 godinu
long_2022$Y <- 2022
dim(long_2022)

## [1] 465696      7

long_2022 <- long_2022 %>% 
  filter(Nivo==3) %>% 
  filter(POL != "Ukupno") %>%
  filter (STAROST != "UKUPNO")
dim (long_2022)

## [1] 265440      7

# transform it to individual level
# Load necessary library
library(tidyr)

# Transform the dataset by expanding rows based on the 'value' column
long_2022_indiv <- long_2022 %>%
  uncount(value)

# Print the transformed dataset
print(long_2022_indiv)

## # A tibble: 24,592 × 6
##     Nivo Opcina         STAROST POL   `iz/u`                Y
##    <dbl> <chr>          <chr>   <chr> <chr>             <dbl>
##  1     3 BOSANSKA KRUPA 0-4     M     BOSANSKI PETROVAC  2022
##  2     3 BOSANSKA KRUPA 0-4     M     BUŽIM              2022
##  3     3 BOSANSKA KRUPA 0-4     M     CAZIN              2022
##  4     3 BOSANSKA KRUPA 0-4     M     CAZIN              2022
##  5     3 BOSANSKA KRUPA 0-4     M     CAZIN              2022
##  6     3 BOSANSKA KRUPA 0-4     M     SANSKI MOST        2022
##  7     3 BOSANSKA KRUPA 0-4     M     ZENICA             2022
##  8     3 BOSANSKA KRUPA 0-4     M     ZENICA             2022
##  9     3 BOSANSKA KRUPA 0-4     M     INO                2022
## 10     3 BOSANSKA KRUPA 0-4     Ž     CAZIN              2022
## # ℹ 24,582 more rows

dim (long_2022_indiv)

## [1] 24592     6

names(long_2022_indiv)

## [1] "Nivo"    "Opcina"  "STAROST" "POL"     "iz/u"    "Y"

#rename
long_2022_indiv <- long_2022_indiv %>% rename(from = Opcina, to = `iz/u`, Pol = POL)

Ponovi za 2020 godinu

#2020 godina sa genderom i dobnim grupama

library(readxl)
X2020_medjuopcinske_Mig <- read_excel("C:/Users/Amra/OneDrive - Direkcija za ekonomsko planiranje/Prijava teme doktorske disertacije/data collection for BiH/!FBiHmedjuopstinskeunutrasnjemigracije/2020_medjuopcinske_Mig.xlsx")
X_2020 <- X2020_medjuopcinske_Mig

library(tidyr)
library(dplyr)
long_2020 <- X_2020 %>% 
pivot_longer(
    cols = "BIHAĆ":"INO", 
    names_to = "iz/u",
    values_to = "value",
    names_repair = "minimal"  #veoma bitno to je kad imamo veoma sitnih razlika i onda kazemo R-u da ih zanemari
  )

# dodaj 2020 godinu

long_2020$Y <- 2020
dim(long_2020)

## [1] 459270      7

long_2020 <- long_2020 %>% 
  filter(Nivo==3) %>% 
  filter(Pol != "Ukupno") %>%
  filter (STAROST != "UKUPNO")
dim (long_2020)

## [1] 255960      7

# transform it to individual level
# Load necessary library
library(tidyr)

# Transform the dataset by expanding rows based on the 'value' column
long_2020_indiv <- long_2020 %>%
  uncount(value)

# Print the transformed dataset
print(long_2020_indiv)

## # A tibble: 19,879 × 6
##     Nivo `KANTON/OPĆINA` STAROST Pol   `iz/u`                 Y
##    <dbl> <chr>           <chr>   <chr> <chr>              <dbl>
##  1     3 BOSANSKA KRUPA  0-4     M     BIHAĆ               2020
##  2     3 BOSANSKA KRUPA  0-4     M     BIHAĆ               2020
##  3     3 BOSANSKA KRUPA  0-4     M     BUŽIM               2020
##  4     3 BOSANSKA KRUPA  0-4     M     CAZIN               2020
##  5     3 BOSANSKA KRUPA  0-4     M     NOVI GRAD SARAJEVO  2020
##  6     3 BOSANSKA KRUPA  0-4     M     INO                 2020
##  7     3 BOSANSKA KRUPA  0-4     M     INO                 2020
##  8     3 BOSANSKA KRUPA  0-4     Ž     BIHAĆ               2020
##  9     3 BOSANSKA KRUPA  0-4     Ž     BIHAĆ               2020
## 10     3 BOSANSKA KRUPA  0-4     Ž     BUŽIM               2020
## # ℹ 19,869 more rows

dim (long_2020_indiv)

## [1] 19879     6

names(long_2020_indiv)

## [1] "Nivo"          "KANTON/OPĆINA" "STAROST"       "Pol"          
## [5] "iz/u"          "Y"

#rename
long_2020_indiv <- long_2020_indiv %>% rename(from = `KANTON/OPĆINA`, to = `iz/u`)
head(long_2020_indiv)

## # A tibble: 6 × 6
##    Nivo from           STAROST Pol   to                     Y
##   <dbl> <chr>          <chr>   <chr> <chr>              <dbl>
## 1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ               2020
## 2     3 BOSANSKA KRUPA 0-4     M     BIHAĆ               2020
## 3     3 BOSANSKA KRUPA 0-4     M     BUŽIM               2020
## 4     3 BOSANSKA KRUPA 0-4     M     CAZIN               2020
## 5     3 BOSANSKA KRUPA 0-4     M     NOVI GRAD SARAJEVO  2020
## 6     3 BOSANSKA KRUPA 0-4     M     INO                 2020

Ponovi za 2019 godinu

#2019 godina sa genderom i dobnim grupama

library(readxl)
X2019_medjuopcinske_Mig <- read_excel("C:/Users/Amra/OneDrive - Direkcija za ekonomsko planiranje/Prijava teme doktorske disertacije/data collection for BiH/!FBiHmedjuopstinskeunutrasnjemigracije/2019_medjuopcinske_Mig.xlsx")
X_2019 <- X2019_medjuopcinske_Mig

library(tidyr)
library(dplyr)
long_2019 <- X_2019 %>% 
pivot_longer(
    cols = "BIHAĆ":"INO", 
    names_to = "iz/u",
    values_to = "value",
    names_repair = "minimal"  #veoma bitno to je kad imamo veoma sitnih razlika i onda kazemo R-u da ih zanemari
  )

# dodaj 2019 godinu
long_2019$Y <- 2019
dim(long_2019)

## [1] 464940      7

long_2019 <- long_2019 %>% 
  filter(Nivo==3) %>% 
  filter(Pol != "Ukupno") %>%
  filter (STAROST != "UKUPNO")
dim (long_2019)

## [1] 259120      7

# transform it to individual level
# Load necessary library
library(tidyr)

# Transform the dataset by expanding rows based on the 'value' column
long_2019_indiv <- long_2019 %>%
  uncount(value)

# Print the transformed dataset
print(long_2019_indiv)

## # A tibble: 22,451 × 6
##     Nivo `KANTON/OPĆINA` STAROST Pol   `iz/u`     Y
##    <dbl> <chr>           <chr>   <chr> <chr>  <dbl>
##  1     3 BOSANSKA KRUPA  0-4     M     BIHAĆ   2019
##  2     3 BOSANSKA KRUPA  0-4     M     CAZIN   2019
##  3     3 BOSANSKA KRUPA  0-4     M     INO     2019
##  4     3 BOSANSKA KRUPA  0-4     M     INO     2019
##  5     3 BOSANSKA KRUPA  0-4     M     INO     2019
##  6     3 BOSANSKA KRUPA  0-4     Ž     MOSTAR  2019
##  7     3 BOSANSKA KRUPA  0-4     Ž     INO     2019
##  8     3 BOSANSKA KRUPA  0-4     Ž     INO     2019
##  9     3 BOSANSKA KRUPA  5-9     M     BIHAĆ   2019
## 10     3 BOSANSKA KRUPA  5-9     M     KLJUČ   2019
## # ℹ 22,441 more rows

dim (long_2019_indiv)

## [1] 22451     6

names(long_2019_indiv)

## [1] "Nivo"          "KANTON/OPĆINA" "STAROST"       "Pol"          
## [5] "iz/u"          "Y"

#rename
long_2019_indiv <- long_2019_indiv %>% rename(from = `KANTON/OPĆINA`, to = `iz/u`)
head(long_2019_indiv)

## # A tibble: 6 × 6
##    Nivo from           STAROST Pol   to         Y
##   <dbl> <chr>          <chr>   <chr> <chr>  <dbl>
## 1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2019
## 2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2019
## 3     3 BOSANSKA KRUPA 0-4     M     INO     2019
## 4     3 BOSANSKA KRUPA 0-4     M     INO     2019
## 5     3 BOSANSKA KRUPA 0-4     M     INO     2019
## 6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2019

Ponovi za 2018 godinu

#2018 godina sa genderom i dobnim grupama

library(readxl)
X2018_medjuopcinske_Mig <- read_excel("C:/Users/Amra/OneDrive - Direkcija za ekonomsko planiranje/Prijava teme doktorske disertacije/data collection for BiH/!FBiHmedjuopstinskeunutrasnjemigracije/2019_medjuopcinske_Mig.xlsx")
X_2018 <- X2018_medjuopcinske_Mig

library(tidyr)
library(dplyr)
long_2018 <- X_2018 %>% 
pivot_longer(
    cols = "BIHAĆ":"INO", 
    names_to = "iz/u",
    values_to = "value",
    names_repair = "minimal"  #veoma bitno to je kad imamo veoma sitnih razlika i onda kazemo R-u da ih zanemari
  )

# dodaj 2018 godinu
long_2018$Y <- 2018
dim(long_2018)

## [1] 464940      7

long_2018 <- long_2018 %>% 
  filter(Nivo==3) %>% 
  filter(Pol != "Ukupno") %>%
  filter (STAROST != "UKUPNO")
dim (long_2018)

## [1] 259120      7

# transform it to individual level
# Load necessary library
library(tidyr)

# Transform the dataset by expanding rows based on the 'value' column
long_2018_indiv <- long_2018 %>%
  uncount(value)

# Print the transformed dataset
print(long_2018_indiv)

## # A tibble: 22,451 × 6
##     Nivo `KANTON/OPĆINA` STAROST Pol   `iz/u`     Y
##    <dbl> <chr>           <chr>   <chr> <chr>  <dbl>
##  1     3 BOSANSKA KRUPA  0-4     M     BIHAĆ   2018
##  2     3 BOSANSKA KRUPA  0-4     M     CAZIN   2018
##  3     3 BOSANSKA KRUPA  0-4     M     INO     2018
##  4     3 BOSANSKA KRUPA  0-4     M     INO     2018
##  5     3 BOSANSKA KRUPA  0-4     M     INO     2018
##  6     3 BOSANSKA KRUPA  0-4     Ž     MOSTAR  2018
##  7     3 BOSANSKA KRUPA  0-4     Ž     INO     2018
##  8     3 BOSANSKA KRUPA  0-4     Ž     INO     2018
##  9     3 BOSANSKA KRUPA  5-9     M     BIHAĆ   2018
## 10     3 BOSANSKA KRUPA  5-9     M     KLJUČ   2018
## # ℹ 22,441 more rows

dim (long_2018_indiv)

## [1] 22451     6

names(long_2018_indiv)

## [1] "Nivo"          "KANTON/OPĆINA" "STAROST"       "Pol"          
## [5] "iz/u"          "Y"

#rename
long_2018_indiv <- long_2018_indiv %>% rename(from = `KANTON/OPĆINA`, to = `iz/u`)
head(long_2018_indiv)

## # A tibble: 6 × 6
##    Nivo from           STAROST Pol   to         Y
##   <dbl> <chr>          <chr>   <chr> <chr>  <dbl>
## 1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2018
## 2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2018
## 3     3 BOSANSKA KRUPA 0-4     M     INO     2018
## 4     3 BOSANSKA KRUPA 0-4     M     INO     2018
## 5     3 BOSANSKA KRUPA 0-4     M     INO     2018
## 6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2018

Uvezi sve baze da budu jedna ispod druge

sve.mig <- bind_rows(long_2018_indiv,long_2019_indiv,long_2020_indiv, long_2021_indiv, long_2022_indiv)
head (sve.mig)

## # A tibble: 6 × 6
##    Nivo from           STAROST Pol   to         Y
##   <dbl> <chr>          <chr>   <chr> <chr>  <dbl>
## 1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2018
## 2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2018
## 3     3 BOSANSKA KRUPA 0-4     M     INO     2018
## 4     3 BOSANSKA KRUPA 0-4     M     INO     2018
## 5     3 BOSANSKA KRUPA 0-4     M     INO     2018
## 6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2018

dodaj Ethnic.majority.iz i Ethnic.majority.u

Ethnic specificity from one database to this sve.mig database

#### import database with ethnic majrity

medjuop_migracijeFBiH_2018_2019 <- read_excel("C:/Users/Amra/OneDrive - Direkcija za ekonomsko planiranje/Prijava teme doktorske disertacije/data collection for BiH/medjuop.migracijeFBiH_2018-2019.xlsx", 
                                                sheet = "2018-2019", skip = 1)
wide <- medjuop_migracijeFBiH_2018_2019

transform from wide to long

library(tidyr)
long <- wide %>% 
pivot_longer(
    cols = "BIHAĆ":"INO", 
    names_to = "iz/u",
    values_to = "value",
    names_repair = "minimal"  #veoma bitno to je kad imamo veoma sitnih razlika i onda kazemo R-u da ih zanemari
  )

####inspect long and rename variables

names(long)[2] <- "from"  # Rename the second column
names(long)[3] <- "to"   # Rename the third column
names (long)

## [1] "Majority" "from"     "to"       "value"

# Create a data frame for the join
join_df <- long %>%
  select(from, Majority) %>%
  distinct()  # Ensure unique pairs of iz and Majority

# Perform a left join to match Majority based on u
sve.mig.ethnic <- sve.mig %>%
  left_join(join_df, by = c("from" = "from"))

print (sve.mig.ethnic)

## # A tibble: 110,807 × 7
##     Nivo from           STAROST Pol   to         Y Majority
##    <dbl> <chr>          <chr>   <chr> <chr>  <dbl> <chr>   
##  1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2018 Bosniak 
##  2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2018 Bosniak 
##  3     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak 
##  4     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak 
##  5     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak 
##  6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2018 Bosniak 
##  7     3 BOSANSKA KRUPA 0-4     Ž     INO     2018 Bosniak 
##  8     3 BOSANSKA KRUPA 0-4     Ž     INO     2018 Bosniak 
##  9     3 BOSANSKA KRUPA 5-9     M     BIHAĆ   2018 Bosniak 
## 10     3 BOSANSKA KRUPA 5-9     M     KLJUČ   2018 Bosniak 
## # ℹ 110,797 more rows

# Rename the new column
sve.mig.ethnic <- sve.mig.ethnic %>%
  rename(Majority_from = Majority)
  
# View the updated data frame
head(sve.mig.ethnic)

## # A tibble: 6 × 7
##    Nivo from           STAROST Pol   to         Y Majority_from
##   <dbl> <chr>          <chr>   <chr> <chr>  <dbl> <chr>        
## 1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2018 Bosniak      
## 2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2018 Bosniak      
## 3     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak      
## 4     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak      
## 5     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak      
## 6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2018 Bosniak

Now ethnic majority for destination municipalities with programing

# Initialize the new column with NA
sve.mig.ethnic$Majority_to <- NA
print(sve.mig)

## # A tibble: 110,807 × 6
##     Nivo from           STAROST Pol   to         Y
##    <dbl> <chr>          <chr>   <chr> <chr>  <dbl>
##  1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2018
##  2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2018
##  3     3 BOSANSKA KRUPA 0-4     M     INO     2018
##  4     3 BOSANSKA KRUPA 0-4     M     INO     2018
##  5     3 BOSANSKA KRUPA 0-4     M     INO     2018
##  6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2018
##  7     3 BOSANSKA KRUPA 0-4     Ž     INO     2018
##  8     3 BOSANSKA KRUPA 0-4     Ž     INO     2018
##  9     3 BOSANSKA KRUPA 5-9     M     BIHAĆ   2018
## 10     3 BOSANSKA KRUPA 5-9     M     KLJUČ   2018
## # ℹ 110,797 more rows

#add from RS, BD, INO
dim (sve.mig.ethnic)

## [1] 110807      8

new_rows <- data.frame(
  Nivo = c(2, 2, 2),
  from = c("REPUBLIKA SRPSKA", "BRČKO DISTRIKT BOSNE I HERCEGOVINE", "INO"),
  STAROST = c(NA, NA, NA),
  Pol = c(NA, NA, NA),
  to = c(NA, NA, NA),  
  Y = c(NA, NA, NA),
  Majority_from = c("Serb", "None", "None"),
  Majority_to = c(NA, NA, NA)
)

# Append the new rows to the existing dataset
sve.mig.ethnic <- rbind(sve.mig.ethnic, new_rows)

dim(sve.mig.ethnic)

## [1] 110810      8

# Loop through the data frame
for (i in 1:nrow(sve.mig.ethnic)) {
  if (sve.mig.ethnic$to[i] %in% sve.mig.ethnic$from) {
    sve.mig.ethnic$Majority_to[i] <- sve.mig.ethnic$Majority_from[which(sve.mig.ethnic$from == sve.mig.ethnic$to[i])]
  }
}
print (sve.mig.ethnic)

## # A tibble: 110,810 × 8
##     Nivo from           STAROST Pol   to         Y Majority_from Majority_to
##    <dbl> <chr>          <chr>   <chr> <chr>  <dbl> <chr>         <chr>      
##  1     3 BOSANSKA KRUPA 0-4     M     BIHAĆ   2018 Bosniak       Bosniak    
##  2     3 BOSANSKA KRUPA 0-4     M     CAZIN   2018 Bosniak       Bosniak    
##  3     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak       None       
##  4     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak       None       
##  5     3 BOSANSKA KRUPA 0-4     M     INO     2018 Bosniak       None       
##  6     3 BOSANSKA KRUPA 0-4     Ž     MOSTAR  2018 Bosniak       None       
##  7     3 BOSANSKA KRUPA 0-4     Ž     INO     2018 Bosniak       None       
##  8     3 BOSANSKA KRUPA 0-4     Ž     INO     2018 Bosniak       None       
##  9     3 BOSANSKA KRUPA 5-9     M     BIHAĆ   2018 Bosniak       Bosniak    
## 10     3 BOSANSKA KRUPA 5-9     M     KLJUČ   2018 Bosniak       <NA>       
## # ℹ 110,800 more rows

dim(sve.mig.ethnic)

## [1] 110810      8

sve.mig.ethnic$ID <- 1:110810

Now make a model

#rename
sve.mig.ethnic <- sve.mig.ethnic %>% 
  rename (Level=Nivo, Age.int = STAROST, Sex=Pol)
str(sve.mig.ethnic)

## tibble [110,810 × 9] (S3: tbl_df/tbl/data.frame)
##  $ Level        : num [1:110810] 3 3 3 3 3 3 3 3 3 3 ...
##  $ from         : chr [1:110810] "BOSANSKA KRUPA" "BOSANSKA KRUPA" "BOSANSKA KRUPA" "BOSANSKA KRUPA" ...
##  $ Age.int      : chr [1:110810] "0-4" "0-4" "0-4" "0-4" ...
##  $ Sex          : chr [1:110810] "M" "M" "M" "M" ...
##  $ to           : chr [1:110810] "BIHAĆ" "CAZIN" "INO" "INO" ...
##  $ Y            : num [1:110810] 2018 2018 2018 2018 2018 ...
##  $ Majority_from: chr [1:110810] "Bosniak" "Bosniak" "Bosniak" "Bosniak" ...
##  $ Majority_to  : chr [1:110810] "Bosniak" "Bosniak" "None" "None" ...
##  $ ID           : int [1:110810] 1 2 3 4 5 6 7 8 9 10 ...

ismirov komentar mail

Ako imas mikro informacije (osobe ili domacinstva) onda mozes da estimiras LOGIT model. M Opcija 1: dependent variable=1 ako se osoba preselila iz “group has same ethnical majority as the origin group” (ovdje koristim grupu jer nemas podatke po opstinama). Dvije napomene. U model dodaj kontrole za “socio-economics (najvaznije etnucku pripadnost)” i zadrzi samo observacije osoba koje su se preselile. Opcija 2: estimiraj model iz opcije 1 za razlicite etnicke grupe: bosnjake, srbe, i hrvate.

Migration summary

table (sve.mig.ethnic$Majority_from, sve.mig.ethnic$Majority_to)

##          
##           Bosniak Croat  None  Serb
##   Bosniak   64977  3518 11816  9738
##   Croat      3345  3033  3799   889
##   None       1805  1648   896   863
##   Serb        231    33   149  1087

Uradicu logic model ismir preporuka

sve.mig.ethnic$same_majority <- ifelse(sve.mig.ethnic$Majority_from == sve.mig.ethnic$Majority_to, 1, 0)

other age intervals

# Load necessary libraries
library(dplyr)

# Recategorize Age.int into broader intervals
sve.mig.ethnic <- sve.mig.ethnic %>%
  mutate(Age_group = case_when(
    Age.int %in% c("0-4", "5-9", "10-14","15-19") ~ "0-19",
    Age.int %in% c("20-24","25-29", "30-34") ~ "20-34",
    Age.int %in% c("35-39", "40-44", "45-49") ~ "35-49",
    Age.int %in% c("50-54", "55-59", "60-64") ~ "50-64",
    Age.int %in% c("65-69", "70-74", "75-79", "80-84", "85+") ~ "65+",
    TRUE ~ NA_character_
  ))

#keep only above 20 as instructed by ismir
sve.mig.ethnic.above19 <- sve.mig.ethnic %>% filter (Age_group!="0-19")

# Re-run the logistic regression with the new age groups
model2 <- glm(same_majority ~ Age_group + Sex + Y+ from, 
              family = binomial(link = "logit"), data = sve.mig.ethnic.above19)

# Summary of the new model
summary(model2)

## 
## Call:
## glm(formula = same_majority ~ Age_group + Sex + Y + from, family = binomial(link = "logit"), 
##     data = sve.mig.ethnic.above19)
## 
## Coefficients:
##                            Estimate Std. Error z value Pr(>|z|)    
## (Intercept)               -9.773060  11.411441  -0.856 0.391762    
## Age_group35-49            -0.508867   0.020129 -25.280  < 2e-16 ***
## Age_group50-64            -0.661776   0.024489 -27.024  < 2e-16 ***
## Age_group65+              -1.219250   0.027207 -44.814  < 2e-16 ***
## SexŽ                       0.488369   0.016719  29.211  < 2e-16 ***
## Y                          0.005457   0.005649   0.966 0.334066    
## fromBIHAĆ                 -0.873578   0.095169  -9.179  < 2e-16 ***
## fromBOSANSKA KRUPA        -0.925852   0.099983  -9.260  < 2e-16 ***
## fromBOSANSKI PETROVAC     -0.136367   0.124492  -1.095 0.273347    
## fromBOSANSKO GRAHOVO      -0.148060   0.187772  -0.789 0.430398    
## fromBREZA                  0.971818   0.157487   6.171 6.80e-10 ***
## fromBUGOJNO               -0.785541   0.101220  -7.761 8.44e-15 ***
## fromBUSOVAČA              -3.399470   0.156530 -21.718  < 2e-16 ***
## fromBUŽIM                 -0.559961   0.112146  -4.993 5.94e-07 ***
## fromCAZIN                 -1.009569   0.094324 -10.703  < 2e-16 ***
## fromČAPLJINA              -1.913066   0.113131 -16.910  < 2e-16 ***
## fromČELIĆ                 -1.783451   0.128609 -13.867  < 2e-16 ***
## fromČITLUK                -1.059280   0.120711  -8.775  < 2e-16 ***
## fromDOBOJ ISTOK           -0.963388   0.127447  -7.559 4.06e-14 ***
## fromDOBOJ JUG             -0.578888   0.150086  -3.857 0.000115 ***
## fromDOBRETIĆI             -2.862802   0.493675  -5.799 6.67e-09 ***
## fromDOMALJEVAC/ŠAMAC      -1.800744   0.206727  -8.711  < 2e-16 ***
## fromDONJI VAKUF           -0.222496   0.125290  -1.776 0.075759 .  
## fromDRVAR                  0.575689   0.142434   4.042 5.30e-05 ***
## fromFOČA                   0.044995   0.191396   0.235 0.814139    
## fromFOJNICA               -1.121417   0.125039  -8.969  < 2e-16 ***
## fromGLAMOČ                -2.660987   0.184809 -14.399  < 2e-16 ***
## fromGORAŽDE               -0.567931   0.101017  -5.622 1.89e-08 ***
## fromGORNJI VAKUF-USKOPLJE -0.562371   0.115349  -4.875 1.09e-06 ***
## fromGRAČANICA             -0.626988   0.101410  -6.183 6.30e-10 ***
## fromGRADAČAC              -1.286781   0.104969 -12.259  < 2e-16 ***
## fromGRUDE                 -0.952303   0.139709  -6.816 9.34e-12 ***
## fromHADŽIĆI                0.273093   0.117427   2.326 0.020037 *  
## fromILIDŽA                 0.258437   0.090404   2.859 0.004254 ** 
## fromILIJAŠ                 0.183944   0.109688   1.677 0.093547 .  
## fromJABLANICA             -1.016977   0.133053  -7.643 2.12e-14 ***
## fromJAJCE                 -2.163796   0.121101 -17.868  < 2e-16 ***
## fromKAKANJ                 0.255589   0.110556   2.312 0.020786 *  
## fromKALESIJA              -0.845514   0.099051  -8.536  < 2e-16 ***
## fromKISELJAK              -3.520519   0.151284 -23.271  < 2e-16 ***
## fromKLADANJ                0.113775   0.124198   0.916 0.359626    
## fromKONJIC                -0.601529   0.108707  -5.533 3.14e-08 ***
## fromKREŠEVO               -1.806882   0.173641 -10.406  < 2e-16 ***
## fromKUPRES                -2.703265   0.224256 -12.054  < 2e-16 ***
## fromLIVNO                 -2.770679   0.146119 -18.962  < 2e-16 ***
## fromLUKAVAC               -0.017383   0.104357  -0.167 0.867708    
## fromLJUBUŠKI              -1.483737   0.118963 -12.472  < 2e-16 ***
## fromMAGLAJ                -0.974966   0.108203  -9.011  < 2e-16 ***
## fromMOSTAR                -2.793663   0.095779 -29.168  < 2e-16 ***
## fromNEUM                  -1.980464   0.194933 -10.160  < 2e-16 ***
## fromNOVI GRAD SARAJEVO     0.265274   0.086984   3.050 0.002291 ** 
## fromNOVI TRAVNIK          -0.736883   0.111046  -6.636 3.23e-11 ***
## fromNOVO SARAJEVO          0.551792   0.088917   6.206 5.45e-10 ***
## fromODŽAK                 -3.997316   0.222748 -17.945  < 2e-16 ***
## fromOLOVO                  1.080579   0.148379   7.283 3.27e-13 ***
## fromORAŠJE                -3.198877   0.187849 -17.029  < 2e-16 ***
## fromPALE                   1.591515   0.476553   3.340 0.000839 ***
## fromPOSUŠJE               -1.226475   0.147059  -8.340  < 2e-16 ***
## fromPROZOR                -3.209364   0.163825 -19.590  < 2e-16 ***
## fromRAVNO                 -1.794998   0.247595  -7.250 4.18e-13 ***
## fromSANSKI MOST           -1.887805   0.101123 -18.668  < 2e-16 ***
## fromSAPNA                 -0.800534   0.120128  -6.664 2.66e-11 ***
## fromSARAJEVO CENTAR        0.546243   0.090889   6.010 1.86e-09 ***
## fromSARAJEVO STARI GRAD    1.013371   0.101700   9.964  < 2e-16 ***
## fromSREBRENIK             -0.909496   0.099318  -9.157  < 2e-16 ***
## fromSTOLAC                -2.331330   0.134439 -17.341  < 2e-16 ***
## fromŠIROKI BRIJEG         -1.451696   0.111680 -12.999  < 2e-16 ***
## fromTEOČAK                -1.291384   0.141337  -9.137  < 2e-16 ***
## fromTEŠANJ                -1.342465   0.099890 -13.439  < 2e-16 ***
## fromTOMISLAVGRAD          -1.962212   0.135650 -14.465  < 2e-16 ***
## fromTRAVNIK               -0.812235   0.095905  -8.469  < 2e-16 ***
## fromTRNOVO                 1.434286   0.209735   6.839 8.00e-12 ***
## fromTUZLA                 -0.736229   0.089252  -8.249  < 2e-16 ***
## fromUSORA                 -3.338471   0.280520 -11.901  < 2e-16 ***
## fromVAREŠ                  0.120687   0.140774   0.857 0.391272    
## fromVELIKA KLADUŠA        -0.989105   0.099850  -9.906  < 2e-16 ***
## fromVISOKO                 0.181501   0.108448   1.674 0.094205 .  
## fromVITEZ                 -4.481120   0.183623 -24.404  < 2e-16 ***
## fromVOGOŠĆA                0.441838   0.103958   4.250 2.14e-05 ***
## fromZAVIDOVIĆI            -0.586802   0.103035  -5.695 1.23e-08 ***
## fromZENICA                -0.760239   0.089683  -8.477  < 2e-16 ***
## fromŽEPČE                 -4.160610   0.171535 -24.255  < 2e-16 ***
## fromŽIVINICE              -0.551747   0.096741  -5.703 1.17e-08 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 113551  on 86371  degrees of freedom
## Residual deviance:  92061  on 86289  degrees of freedom
##   (2377 observations deleted due to missingness)
## AIC: 92227
## 
## Number of Fisher Scoring iterations: 5

model2a <- glm(same_majority ~ Age_group + Sex + Majority_from + factor(Y), 
              family = binomial(link = "logit"), data = sve.mig.ethnic.above19)
summary(model2a)

## 
## Call:
## glm(formula = same_majority ~ Age_group + Sex + Majority_from + 
##     factor(Y), family = binomial(link = "logit"), data = sve.mig.ethnic.above19)
## 
## Coefficients:
##                      Estimate Std. Error z value Pr(>|z|)    
## (Intercept)         9.611e-01  2.180e-02  44.082  < 2e-16 ***
## Age_group35-49     -4.007e-01  1.895e-02 -21.137  < 2e-16 ***
## Age_group50-64     -4.951e-01  2.293e-02 -21.594  < 2e-16 ***
## Age_group65+       -9.916e-01  2.529e-02 -39.212  < 2e-16 ***
## SexŽ                4.019e-01  1.580e-02  25.447  < 2e-16 ***
## Majority_fromCroat -2.010e+00  2.587e-02 -77.721  < 2e-16 ***
## Majority_fromNone  -2.424e+00  4.078e-02 -59.428  < 2e-16 ***
## Majority_fromSerb   3.528e-01  6.682e-02   5.280 1.29e-07 ***
## factor(Y)2019       3.188e-14  2.404e-02   0.000    1.000    
## factor(Y)2020      -1.952e-01  2.440e-02  -8.000 1.25e-15 ***
## factor(Y)2021      -1.459e-02  2.421e-02  -0.603    0.547    
## factor(Y)2022       2.519e-02  2.413e-02   1.044    0.297    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 113551  on 86371  degrees of freedom
## Residual deviance: 100185  on 86360  degrees of freedom
##   (2377 observations deleted due to missingness)
## AIC: 100209
## 
## Number of Fisher Scoring iterations: 4

texreg::screenreg(list(model2, model2a))

## 
## =======================================================
##                            Model 1        Model 2      
## -------------------------------------------------------
## (Intercept)                    -9.77           0.96 ***
##                               (11.41)         (0.02)   
## Age_group35-49                 -0.51 ***      -0.40 ***
##                                (0.02)         (0.02)   
## Age_group50-64                 -0.66 ***      -0.50 ***
##                                (0.02)         (0.02)   
## Age_group65+                   -1.22 ***      -0.99 ***
##                                (0.03)         (0.03)   
## SexŽ                            0.49 ***       0.40 ***
##                                (0.02)         (0.02)   
## Y                               0.01                   
##                                (0.01)                  
## fromBIHAĆ                      -0.87 ***               
##                                (0.10)                  
## fromBOSANSKA KRUPA             -0.93 ***               
##                                (0.10)                  
## fromBOSANSKI PETROVAC          -0.14                   
##                                (0.12)                  
## fromBOSANSKO GRAHOVO           -0.15                   
##                                (0.19)                  
## fromBREZA                       0.97 ***               
##                                (0.16)                  
## fromBUGOJNO                    -0.79 ***               
##                                (0.10)                  
## fromBUSOVAČA                   -3.40 ***               
##                                (0.16)                  
## fromBUŽIM                      -0.56 ***               
##                                (0.11)                  
## fromCAZIN                      -1.01 ***               
##                                (0.09)                  
## fromČAPLJINA                   -1.91 ***               
##                                (0.11)                  
## fromČELIĆ                      -1.78 ***               
##                                (0.13)                  
## fromČITLUK                     -1.06 ***               
##                                (0.12)                  
## fromDOBOJ ISTOK                -0.96 ***               
##                                (0.13)                  
## fromDOBOJ JUG                  -0.58 ***               
##                                (0.15)                  
## fromDOBRETIĆI                  -2.86 ***               
##                                (0.49)                  
## fromDOMALJEVAC/ŠAMAC           -1.80 ***               
##                                (0.21)                  
## fromDONJI VAKUF                -0.22                   
##                                (0.13)                  
## fromDRVAR                       0.58 ***               
##                                (0.14)                  
## fromFOČA                        0.04                   
##                                (0.19)                  
## fromFOJNICA                    -1.12 ***               
##                                (0.13)                  
## fromGLAMOČ                     -2.66 ***               
##                                (0.18)                  
## fromGORAŽDE                    -0.57 ***               
##                                (0.10)                  
## fromGORNJI VAKUF-USKOPLJE      -0.56 ***               
##                                (0.12)                  
## fromGRAČANICA                  -0.63 ***               
##                                (0.10)                  
## fromGRADAČAC                   -1.29 ***               
##                                (0.10)                  
## fromGRUDE                      -0.95 ***               
##                                (0.14)                  
## fromHADŽIĆI                     0.27 *                 
##                                (0.12)                  
## fromILIDŽA                      0.26 **                
##                                (0.09)                  
## fromILIJAŠ                      0.18                   
##                                (0.11)                  
## fromJABLANICA                  -1.02 ***               
##                                (0.13)                  
## fromJAJCE                      -2.16 ***               
##                                (0.12)                  
## fromKAKANJ                      0.26 *                 
##                                (0.11)                  
## fromKALESIJA                   -0.85 ***               
##                                (0.10)                  
## fromKISELJAK                   -3.52 ***               
##                                (0.15)                  
## fromKLADANJ                     0.11                   
##                                (0.12)                  
## fromKONJIC                     -0.60 ***               
##                                (0.11)                  
## fromKREŠEVO                    -1.81 ***               
##                                (0.17)                  
## fromKUPRES                     -2.70 ***               
##                                (0.22)                  
## fromLIVNO                      -2.77 ***               
##                                (0.15)                  
## fromLUKAVAC                    -0.02                   
##                                (0.10)                  
## fromLJUBUŠKI                   -1.48 ***               
##                                (0.12)                  
## fromMAGLAJ                     -0.97 ***               
##                                (0.11)                  
## fromMOSTAR                     -2.79 ***               
##                                (0.10)                  
## fromNEUM                       -1.98 ***               
##                                (0.19)                  
## fromNOVI GRAD SARAJEVO          0.27 **                
##                                (0.09)                  
## fromNOVI TRAVNIK               -0.74 ***               
##                                (0.11)                  
## fromNOVO SARAJEVO               0.55 ***               
##                                (0.09)                  
## fromODŽAK                      -4.00 ***               
##                                (0.22)                  
## fromOLOVO                       1.08 ***               
##                                (0.15)                  
## fromORAŠJE                     -3.20 ***               
##                                (0.19)                  
## fromPALE                        1.59 ***               
##                                (0.48)                  
## fromPOSUŠJE                    -1.23 ***               
##                                (0.15)                  
## fromPROZOR                     -3.21 ***               
##                                (0.16)                  
## fromRAVNO                      -1.79 ***               
##                                (0.25)                  
## fromSANSKI MOST                -1.89 ***               
##                                (0.10)                  
## fromSAPNA                      -0.80 ***               
##                                (0.12)                  
## fromSARAJEVO CENTAR             0.55 ***               
##                                (0.09)                  
## fromSARAJEVO STARI GRAD         1.01 ***               
##                                (0.10)                  
## fromSREBRENIK                  -0.91 ***               
##                                (0.10)                  
## fromSTOLAC                     -2.33 ***               
##                                (0.13)                  
## fromŠIROKI BRIJEG              -1.45 ***               
##                                (0.11)                  
## fromTEOČAK                     -1.29 ***               
##                                (0.14)                  
## fromTEŠANJ                     -1.34 ***               
##                                (0.10)                  
## fromTOMISLAVGRAD               -1.96 ***               
##                                (0.14)                  
## fromTRAVNIK                    -0.81 ***               
##                                (0.10)                  
## fromTRNOVO                      1.43 ***               
##                                (0.21)                  
## fromTUZLA                      -0.74 ***               
##                                (0.09)                  
## fromUSORA                      -3.34 ***               
##                                (0.28)                  
## fromVAREŠ                       0.12                   
##                                (0.14)                  
## fromVELIKA KLADUŠA             -0.99 ***               
##                                (0.10)                  
## fromVISOKO                      0.18                   
##                                (0.11)                  
## fromVITEZ                      -4.48 ***               
##                                (0.18)                  
## fromVOGOŠĆA                     0.44 ***               
##                                (0.10)                  
## fromZAVIDOVIĆI                 -0.59 ***               
##                                (0.10)                  
## fromZENICA                     -0.76 ***               
##                                (0.09)                  
## fromŽEPČE                      -4.16 ***               
##                                (0.17)                  
## fromŽIVINICE                   -0.55 ***               
##                                (0.10)                  
## Majority_fromCroat                            -2.01 ***
##                                               (0.03)   
## Majority_fromNone                             -2.42 ***
##                                               (0.04)   
## Majority_fromSerb                              0.35 ***
##                                               (0.07)   
## factor(Y)2019                                  0.00    
##                                               (0.02)   
## factor(Y)2020                                 -0.20 ***
##                                               (0.02)   
## factor(Y)2021                                 -0.01    
##                                               (0.02)   
## factor(Y)2022                                  0.03    
##                                               (0.02)   
## -------------------------------------------------------
## AIC                         92226.82      100209.44    
## BIC                         93004.24      100321.83    
## Log Likelihood             -46030.41      -50092.72    
## Deviance                    92060.82      100185.44    
## Num. obs.                   86372          86372       
## =======================================================
## *** p < 0.001; ** p < 0.01; * p < 0.05

This is a logistic regression model (GLM) with a binomial family, which predicts the probability of migration from one ethnic majority area to another based on factors like age, sex, ethnic majority from which the person is moving, and the year of migration (Y). The outcome variable is same_majority, indicating whether a person moved between areas of the same ethnic majority.

Key interpretations:

Intercept (0.9611): The log-odds of moving to a municipality with the same ethnic majority for the reference categories (age 20-34, male, from Bosniak majority, in 2018). A positive intercept suggests a higher likelihood of such a move for the reference group.
Age Groups:
- Compared to the reference age group (20-34):
  - Age 25-49 (-0.4007): There is a decrease in the log-odds of moving to the same ethnic majority area for people aged 25-49.
  - Age 50-64 (-0.4951): Even more negative, suggesting a further reduced likelihood for people aged 50-64.
  - Age 65+ (-0.9916): This group has the largest decrease, indicating significantly lower odds of moving between same-ethnic-majority areas for older people.
Sex (Female = 0.4019): Females have higher odds of moving between areas of the same ethnic majority compared to males (reference category).
Majority_from:
- Croat (-2.010): Individuals from Croat-majority areas are less likely to move to areas of the same ethnic majority compared to those from Bosniak-majority areas.
- None (-2.424): Those from areas with no ethnic majority have even lower odds of moving to an area of the same majority.
- Serb (0.3528): People from Serb-majority areas have slightly increased odds of moving to an area of the same ethnic majority compared to Bosniak-majority areas.
Year (factor(Y)):
- 2019 (0.000): There is no significant difference in the likelihood of moving to the same ethnic majority area compared to 2018.
- 2020 (-0.1952): Lower odds of moving between areas of the same ethnic majority in 2020 compared to 2018.
- 2021 (-0.0146) and 2022 (0.0252): These years do not show significant differences from 2018.

Statistical Significance:

Most predictors (except the years 2019, 2021, and 2022) are highly significant (p < 0.001), as indicated by the *** symbols, meaning the relationships observed in the model are unlikely to have occurred by chance.

Model Fit:

Null deviance (113551) vs. Residual deviance (100185): The drop in deviance indicates that the model fits better than a null model (one with no predictors).
AIC (100209): This is used for model comparison; lower values indicate better model fit.

Conclusion:

The likelihood of migrating between municipalities with the same ethnic majority decreases with age, especially for older people.
Females are more likely than males to move between such areas.
People from Croat and non-majority areas are much less likely to move to areas of the same ethnic majority compared to those from Bosniak-majority areas, while people from Serb-majority areas are slightly more likely.
There were some small but significant variations by year, especially in 2020 when the likelihood of moving to the same majority area decreased.

Yes, from the logistic regression model, we can calculate the predicted probabilities of migration to a municipality with the same ethnic majority group for each ethnic group, holding other variables constant.

The logistic regression equation is:

\[ P(\text{same_majority} = 1) = \frac{1}{1 + e^{-(\beta_0 + \beta_1X_1 + \dots + \beta_kX_k)}} \]

Where: - \(\beta_0\) is the intercept. - \(\beta_1, \beta_2, \dots\) are the coefficients for the respective variables. - \(X_1, X_2, \dots\) are the values of the predictor variables (like ethnic majority).

Step-by-Step to Calculate Probability:

For Bosniak (reference category):
- The baseline is determined by the intercept \(\beta_0\), as “Bosniak” is the reference category for both Majority_from and Majority_to.
- Predicted log-odds for Bosniak: \(\beta_0 = 2.95665\).
- Probability for Bosniak: \[ P(\text{same majority} | \text{Bosniak}) = \frac{1}{1 + e^{-2.95665}} = 0.9506 \]
- Bosniaks have a 95.06% probability of migrating to another Bosniak-majority municipality.
For Croat:
- Predicted log-odds: \[ \text{log-odds} = \beta_0 + \beta_{\text{Majority_from: Croat}} = 2.95665 - 2.52691 = 0.42974 \]
- Probability for Croat: \[ P(\text{same majority} | \text{Croat}) = \frac{1}{1 + e^{-0.42974}} = 0.6058 \]
- Croats have a 60.58% probability of migrating to another Croat-majority municipality.
For Serb:
- Predicted log-odds: \[ \text{log-odds} = \beta_0 + \beta_{\text{Majority_from: Serb}} = 2.95665 + 3.85851 = 6.81516 \]
- Probability for Serb: \[ P(\text{same majority} | \text{Serb}) = \frac{1}{1 + e^{-6.81516}} \approx 0.999 \]
- Serbs have nearly a 100% probability of migrating to another Serb-majority municipality.
For “None” (no majority):
- Predicted log-odds: \[ \text{log-odds} = \beta_0 + \beta_{\text{Majority_from: None}} = 2.95665 - 3.45323 = -0.49658 \]
- Probability for “None”: \[ P(\text{same majority} | \text{None}) = \frac{1}{1 + e^{0.49658}} = 0.3783 \]
- Individuals from municipalities with no ethnic majority have a 37.83% probability of migrating to a similar municipality.

Summary of Probabilities:

Bosniak: 95.06%
Croat: 60.58%
Serb: ~100%
None (no majority): 37.83%

These probabilities reflect the likelihood of individuals from each ethnic majority group migrating to a municipality with the same ethnic majority. Would you like to explore any of these groups further or include other variables in the calculation?