#— title: “ODS 9. Industria Innovación e Infraestructura: CALCULO DE INDICADOR SINTETICO” author: “Abel Peña” date: “2025-06-24” output: prettydoc::html_pretty: theme: architect highlight: github math: katex —
1. INSTALAR PAQUETES (si es necesario)
# install.packages(c("wbstats","dplyr","tidyverse","magrittr","ggplot2","kableExtra","writexl"))2. CARGAR LIBRERÍAS
# (ya se hace en el chunk de setup)3. OBTENCIÓN DE DATOS DESDE WORLD BANK (ODS 9)
# 3.1 Metadatos y ejemplo puntual
indicadores_disponibles <- wb_indicators(lang = "es")
datos_ejemplo <- wb_data(
indicator = "NV.IND.MANF.ZS", # Manufactura % PIB (ODS 9)
country = "SLV",
start_date = 2010,
end_date = 2024,
lang = "es"
)
# 3.2 Indicadores ODS 9
ods9_indicators <- c(
"NV.IND.MANF.ZS", # Manufactura (% del PIB)
"EG.ELC.ACCS.ZS", # Acceso a electricidad (% población)
"IT.NET.USER.ZS", # Usuarios de Internet (% población)
"GB.XPD.RSDV.GD.ZS" # Gasto en I+D (% del PIB)
)
# 3.3 Países por región + El Salvador
countries_region <- c(
"CAN","USA","MEX", # Norteamérica
"BLZ","GTM","HND","NIC","CRI","PAN", # Centroamérica
"HTI","CUB","JAM","DOM", # Caribe
"COL","VEN","PER","CHL","BOL","ARG","BRA", # Sudamérica
"SLV" # El Salvador
)
# 3.4 Descargar datos ODS9 2010–2024
data_ods9 <- wb_data(
indicator = ods9_indicators,
country = countries_region,
start_date = 2010,
end_date = 2024,
lang = "es"
)
# 3.5 Asignar región
data_ods9 <- data_ods9 %>%
mutate(region = case_when(
iso3c %in% c("CAN","USA","MEX") ~ "Norteamérica",
iso3c %in% c("BLZ","GTM","HND","NIC","CRI","PAN") ~ "Centroamérica",
iso3c %in% c("HTI","CUB","JAM","DOM") ~ "Caribe",
iso3c %in% c("COL","VEN","PER","CHL","BOL","ARG","BRA") ~ "Sudamérica",
iso3c == "SLV" ~ "El Salvador"
))4. RESUMEN ESTADÍSTICO
# Estadísticas descriptivas por región
estad_ods9 <- data_ods9 %>%
group_by(region) %>%
summarize(across(all_of(ods9_indicators),
list(mean = ~mean(.x, na.rm=TRUE),
sd = ~sd(.x, na.rm=TRUE),
min = ~min(.x, na.rm=TRUE),
max = ~max(.x, na.rm=TRUE)),
.names = "{col}_{fn}"))
estad_ods9 %>% kable(format = "html", caption = "Descriptivas ODS9 por región")| region | NV.IND.MANF.ZS_mean | NV.IND.MANF.ZS_sd | NV.IND.MANF.ZS_min | NV.IND.MANF.ZS_max | EG.ELC.ACCS.ZS_mean | EG.ELC.ACCS.ZS_sd | EG.ELC.ACCS.ZS_min | EG.ELC.ACCS.ZS_max | IT.NET.USER.ZS_mean | IT.NET.USER.ZS_sd | IT.NET.USER.ZS_min | IT.NET.USER.ZS_max | GB.XPD.RSDV.GD.ZS_mean | GB.XPD.RSDV.GD.ZS_sd | GB.XPD.RSDV.GD.ZS_min | GB.XPD.RSDV.GD.ZS_max |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Caribe | 13.42832 | 3.9334461 | 7.673248 | 24.47841 | 84.23214 | 24.2925960 | 37.2 | 100 | 48.14750 | 24.33476 | 8.37 | 85.2 | 0.4361631 | 0.0988444 | 0.27192 | 0.60767 |
| Centroamérica | 11.94282 | 4.2783391 | 4.982167 | 19.42673 | 91.82143 | 5.8626256 | 80.1 | 100 | 44.64524 | 20.41581 | 10.00 | 85.4 | 0.1765940 | 0.1658248 | 0.01497 | 0.55627 |
| El Salvador | 15.67616 | 0.7099447 | 14.003302 | 16.38749 | 96.17857 | 2.4826098 | 91.6 | 100 | 38.32143 | 18.51325 | 15.90 | 67.7 | 0.1217500 | 0.0533814 | 0.03397 | 0.18078 |
| Norteamérica | 13.95116 | 4.5087747 | 9.015477 | 21.37379 | 99.81667 | 0.3348947 | 99.0 | 100 | 76.32381 | 17.81400 | 31.10 | 94.6 | 1.6636193 | 1.0947784 | 0.25782 | 3.58623 |
| Sudamérica | 11.92949 | 1.9675176 | 8.525428 | 16.31750 | 97.72551 | 3.3307784 | 88.0 | 100 | 60.42065 | 17.31381 | 22.40 | 94.5 | 0.4791794 | 0.3604816 | 0.05530 | 1.37093 |
# Boxplot: Manufactura % PIB
ggplot(data_ods9, aes(x = region, y = NV.IND.MANF.ZS, fill = region)) +
geom_boxplot() +
labs(title = "Manufactura (% PIB) por región", y = "% PIB") +
theme_minimal()
# Serie temporal: Gasto en I+D
ggplot(data_ods9, aes(x = as.integer(date), y = GB.XPD.RSDV.GD.ZS, color = region)) +
geom_line() +
labs(title = "Gasto en I+D (% PIB) 2010–2024", x = "Año", y = "% PIB") +
theme_minimal()
5. CORRELACIONES TEÓRICAS Y EMPÍRICAS
# Empírica
data_corr <- data_ods9 %>%
select(all_of(ods9_indicators)) %>%
drop_na()
cor_emp_ods9 <- cor(data_corr)
cor_emp_ods9 %>% round(2) %>% kable(caption = "Correlación empírica ODS9")| NV.IND.MANF.ZS | EG.ELC.ACCS.ZS | IT.NET.USER.ZS | GB.XPD.RSDV.GD.ZS | |
|---|---|---|---|---|
| NV.IND.MANF.ZS | 1.00 | -0.08 | -0.42 | -0.25 |
| EG.ELC.ACCS.ZS | -0.08 | 1.00 | 0.67 | 0.42 |
| IT.NET.USER.ZS | -0.42 | 0.67 | 1.00 | 0.54 |
| GB.XPD.RSDV.GD.ZS | -0.25 | 0.42 | 0.54 | 1.00 |
# Teóricas (comentarios):
# - Manufactura ↔ Electricidad (+)
# - I+D ↔ Internet (+)
# - Electricidad ↔ Internet (+)6. ANÁLISIS FACTORIAL PCA
# Escalar datos
pca_data_ods9 <- scale(data_corr)
# PCA
pca_ods9 <- prcomp(pca_data_ods9, center = FALSE, scale. = FALSE)
# Cargas PC1-PC2
df_cargas_ods9 <- as.data.frame(pca_ods9$rotation[,1:2])
colnames(df_cargas_ods9) <- c("PC1","PC2")
df_cargas_ods9 %>% kable(caption = "Cargas factoriales ODS9")| PC1 | PC2 | |
|---|---|---|
| NV.IND.MANF.ZS | -0.3375451 | 0.8705741 |
| EG.ELC.ACCS.ZS | 0.5166514 | 0.4870955 |
| IT.NET.USER.ZS | 0.6055066 | 0.0129801 |
| GB.XPD.RSDV.GD.ZS | 0.5024902 | 0.0683393 |
# Scree Plot
plot(pca_ods9, type = "l", main = "Scree Plot PCA ODS9")
7. CREACIÓN DE ÍNDICE SINTÉTICO ODS9
# Normalización min-max
dat_norm_ods9 <- data_ods9 %>%
mutate(across(all_of(ods9_indicators), ~(. - min(., na.rm=TRUE))/(max(., na.rm=TRUE)-min(., na.rm=TRUE)),
.names = "min_{col}"))
# Pesos iguales
dat_norm_ods9 <- dat_norm_ods9 %>%
rowwise() %>%
mutate(IndexODS9_eq = mean(c_across(starts_with("min_")), na.rm=TRUE)) %>%
ungroup()
# Pesos PCA
pesos_ods9 <- pca_ods9$rotation[,1]
pesos_ods9 <- pesos_ods9 / sum(pesos_ods9)
# Calcular índice con pesos PCA
dat_norm_ods9 <- dat_norm_ods9 %>%
rowwise() %>%
mutate(IndexODS9_pca = sum(pesos_ods9 * c_across(all_of(ods9_indicators)), na.rm=TRUE)) %>%
ungroup()
# Gráfico de serie del índice PCA
ggplot(dat_norm_ods9, aes(x = as.integer(date), y = IndexODS9_pca, color = region)) +
geom_line() +
labs(title = "Índice Sintético ODS9 (PCA)", x = "Año", y = "Índice") +
theme_minimal()
8. EXPORTACIÓN A EXCEL
# Preparar tablas
tabla_resumen_ods9 <- estad_ods9
tabla_indice_ods9 <- dat_norm_ods9 %>% select(iso3c, country, date, IndexODS9_eq, IndexODS9_pca)
write_xlsx(
list(
"Descriptivas_ODS9" = tabla_resumen_ods9,
"Cargas_PCA_ODS9" = df_cargas_ods9,
"Indices_ODS9" = tabla_indice_ods9
),
path = "Resultados_ODS9.xlsx"
)