Actividad 1 MOD 1 Generación de Escenarios Futuros con analítica

Actividad sesion 1

Modelo econométrico

Instalar paquetes

#install.packages("shiny")
library(shiny)
#install.packages("WDI")
library(WDI)
#install.packages("wbstats")
library(wbstats)
#install.packages("tidyverse")
library(ggplot2)
#install.packages("plm")
library(plm)
#install.packages("gplots")  
library(gplots)   

## 
## Attaching package: 'gplots'

## The following object is masked from 'package:stats':
## 
##     lowess

Obtener información de 1 país

PIB_MEX <- wb_data(country = "MX", indicator = "NY.GDP.PCAP.CD", 
                   start_date = 1900, end_date = 2025) 

Serie de tiempo

summary(PIB_MEX)

##     iso2c              iso3c             country               date     
##  Length:64          Length:64          Length:64          Min.   :1960  
##  Class :character   Class :character   Class :character   1st Qu.:1976  
##  Mode  :character   Mode  :character   Mode  :character   Median :1992  
##                                                           Mean   :1992  
##                                                           3rd Qu.:2007  
##                                                           Max.   :2023  
##  NY.GDP.PCAP.CD        unit            obs_status          footnote        
##  Min.   :  355.1   Length:64          Length:64          Length:64         
##  1st Qu.: 1427.8   Class :character   Class :character   Class :character  
##  Median : 4006.5   Mode  :character   Mode  :character   Mode  :character  
##  Mean   : 5097.1                                                           
##  3rd Qu.: 8905.4                                                           
##  Max.   :13790.0                                                           
##   last_updated       
##  Min.   :2025-01-28  
##  1st Qu.:2025-01-28  
##  Median :2025-01-28  
##  Mean   :2025-01-28  
##  3rd Qu.:2025-01-28  
##  Max.   :2025-01-28

ggplot(PIB_MEX, aes(x = date, y = NY.GDP.PCAP.CD)) +
  geom_point() + 
  geom_line() + 
  labs(title="PIB per capita en méxico (Current USD)", x="Año",
       y ="Valor")

Obtener información de varios paises

PIB_PANEL <- wb_data(country = c("MX","US","CA" ), indicator = "NY.GDP.PCAP.CD", 
                   start_date = 1900, end_date = 2025) 

summary(PIB_PANEL)

##     iso2c              iso3c             country               date     
##  Length:192         Length:192         Length:192         Min.   :1960  
##  Class :character   Class :character   Class :character   1st Qu.:1976  
##  Mode  :character   Mode  :character   Mode  :character   Median :1992  
##                                                           Mean   :1992  
##                                                           3rd Qu.:2007  
##                                                           Max.   :2023  
##  NY.GDP.PCAP.CD        unit            obs_status          footnote        
##  Min.   :  355.1   Length:192         Length:192         Length:192        
##  1st Qu.: 4059.2   Class :character   Class :character   Class :character  
##  Median :10544.4   Mode  :character   Mode  :character   Mode  :character  
##  Mean   :19152.2                                                           
##  3rd Qu.:29010.1                                                           
##  Max.   :82769.4                                                           
##   last_updated       
##  Min.   :2025-01-28  
##  1st Qu.:2025-01-28  
##  Median :2025-01-28  
##  Mean   :2025-01-28  
##  3rd Qu.:2025-01-28  
##  Max.   :2025-01-28

ggplot(PIB_PANEL, aes(x = date, y = NY.GDP.PCAP.CD, color = iso3c)) +
  geom_point() + 
  geom_line() + 
  labs(title="PIB per capita en Norteamerica (Current USD)", x="Año",
       y ="Valor")

MEGAPIB <- wb_data(country = c("MX","US","CA" ), indicator = c("NY.GDP.PCAP.CD","SP.DYN.LE00.IN"), 
                     start_date = 1900, end_date = 2025) 

summary(MEGAPIB)

##     iso2c              iso3c             country               date     
##  Length:192         Length:192         Length:192         Min.   :1960  
##  Class :character   Class :character   Class :character   1st Qu.:1976  
##  Mode  :character   Mode  :character   Mode  :character   Median :1992  
##                                                           Mean   :1992  
##                                                           3rd Qu.:2007  
##                                                           Max.   :2023  
##                                                                         
##  NY.GDP.PCAP.CD    SP.DYN.LE00.IN 
##  Min.   :  355.1   Min.   :55.02  
##  1st Qu.: 4059.2   1st Qu.:71.11  
##  Median :10544.4   Median :74.36  
##  Mean   :19152.2   Mean   :73.41  
##  3rd Qu.:29010.1   3rd Qu.:77.49  
##  Max.   :82769.4   Max.   :82.22  
##                    NA's   :3

Heterogeneidad

# Heterogeneidad
# Variación entre individuos
plotmeans(NY.GDP.PCAP.CD ~ country,main = "Heterogeneidad entre países", data = MEGAPIB)

## Warning in arrows(x, li, x, pmax(y - gap, li), col = barcol, lwd = lwd, :
## zero-length arrow is of indeterminate angle and so skipped

## Warning in arrows(x, ui, x, pmin(y + gap, ui), col = barcol, lwd = lwd, :
## zero-length arrow is of indeterminate angle and so skipped

# Interpretacion
# Alta Heterogeneidad: Si los puntos (medias) estan muy separados entre países
# Baja Heterogeneidad: Si los puntos (medias) estan cerca uno de otros
# En este caso, EUA y Canadá tienen un PIB per capita mayor que Mexico, mostrando alta heterogeneidad entre países

Modelo de efectos fijos y aleatorios

Paso 1. Convertir los datos a panel

datos_panel<- pdata.frame(PIB_PANEL, index = c("country", "date"))

Modelo de efectos fijos

modeloefcfijos <- plm( NY.GDP.PCAP.CD ~ date, data=datos_panel , model = "within")
summary(modeloefcfijos)

## Oneway (individual) effect Within Model
## 
## Call:
## plm(formula = NY.GDP.PCAP.CD ~ date, data = datos_panel, model = "within")
## 
## Balanced Panel: n = 3, T = 64, N = 192
## 
## Residuals:
##      Min.   1st Qu.    Median   3rd Qu.      Max. 
## -22869.42  -3713.59   -740.79   4417.57  22788.54 
## 
## Coefficients:
##           Estimate Std. Error t-value  Pr(>|t|)    
## date1961    19.689   7891.777  0.0025 0.9980133    
## date1962    93.003   7891.777  0.0118 0.9906159    
## date1963   182.117   7891.777  0.0231 0.9816255    
## date1964   329.256   7891.777  0.0417 0.9667868    
## date1965   493.812   7891.777  0.0626 0.9502057    
## date1966   705.548   7891.777  0.0894 0.9289037    
## date1967   836.074   7891.777  0.1059 0.9157965    
## date1968  1051.287   7891.777  0.1332 0.8942375    
## date1969  1278.661   7891.777  0.1620 0.8715461    
## date1970  1483.079   7891.777  0.1879 0.8512361    
## date1971  1757.600   7891.777  0.2227 0.8241196    
## date1972  2139.145   7891.777  0.2711 0.7867884    
## date1973  2652.616   7891.777  0.3361 0.7373364    
## date1974  3306.205   7891.777  0.4189 0.6759711    
## date1975  3736.686   7891.777  0.4735 0.6366822    
## date1976  4425.604   7891.777  0.5608 0.5759388    
## date1977  4698.806   7891.777  0.5954 0.5526405    
## date1978  5234.634   7891.777  0.6633 0.5083487    
## date1979  6060.354   7891.777  0.7679 0.4439640    
## date1980  7072.576   7891.777  0.8962 0.3718573    
## date1981  8188.133   7891.777  1.0376 0.3014655    
## date1982  7987.390   7891.777  1.0121 0.3134224    
## date1983  8523.654   7891.777  1.0801 0.2821751    
## date1984  9312.706   7891.777  1.1801 0.2402027    
## date1985  9796.257   7891.777  1.2413 0.2167918    
## date1986  9909.818   7891.777  1.2557 0.2115431    
## date1987 10895.002   7891.777  1.3806 0.1698612    
## date1988 12362.836   7891.777  1.5665 0.1197288    
## date1989 13585.668   7891.777  1.7215 0.0876150 .  
## date1990 14316.347   7891.777  1.8141 0.0720442 .  
## date1991 14759.335   7891.777  1.8702 0.0637741 .  
## date1992 14990.000   7891.777  1.8994 0.0597918 .  
## date1993 15667.517   7891.777  1.9853 0.0492832 *  
## date1994 16091.651   7891.777  2.0390 0.0435376 *  
## date1995 15978.167   7891.777  2.0247 0.0450159 *  
## date1996 16773.055   7891.777  2.1254 0.0355067 *  
## date1997 17769.387   7891.777  2.2516 0.0260772 *  
## date1998 18030.354   7891.777  2.2847 0.0240026 *  
## date1999 19236.904   7891.777  2.4376 0.0161811 *  
## date2000 20835.037   7891.777  2.6401 0.0093360 ** 
## date2001 21096.198   7891.777  2.6732 0.0085083 ** 
## date2002 21538.969   7891.777  2.7293 0.0072554 ** 
## date2003 23202.118   7891.777  2.9400 0.0039054 ** 
## date2004 25366.654   7891.777  3.2143 0.0016609 ** 
## date2005 27852.977   7891.777  3.5294 0.0005823 ***
## date2006 30232.924   7891.777  3.8309 0.0002003 ***
## date2007 32408.252   7891.777  4.1066 7.172e-05 ***
## date2008 33394.731   7891.777  4.2316 4.431e-05 ***
## date2009 30291.171   7891.777  3.8383 0.0001950 ***
## date2010 33440.081   7891.777  4.2373 4.333e-05 ***
## date2011 35778.148   7891.777  4.5336 1.331e-05 ***
## date2012 36526.334   7891.777  4.6284 9.027e-06 ***
## date2013 37214.927   7891.777  4.7157 6.286e-06 ***
## date2014 37345.549   7891.777  4.7322 5.866e-06 ***
## date2015 35011.917   7891.777  4.4365 1.971e-05 ***
## date2016 34666.237   7891.777  4.3927 2.348e-05 ***
## date2017 36493.760   7891.777  4.6243 9.182e-06 ***
## date2018 38068.376   7891.777  4.8238 3.990e-06 ***
## date2019 38902.406   7891.777  4.9295 2.543e-06 ***
## date2020 37056.865   7891.777  4.6956 6.833e-06 ***
## date2021 42836.438   7891.777  5.4280 2.815e-07 ***
## date2022 46436.696   7891.777  5.8842 3.387e-08 ***
## date2023 48123.578   7891.777  6.0979 1.218e-08 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Total Sum of Squares:    5.061e+10
## Residual Sum of Squares: 1.1771e+10
## R-Squared:      0.76742
## Adj. R-Squared: 0.64743
## F-statistic: 6.59909 on 63 and 126 DF, p-value: < 2.22e-16

Modelo de efectos aleatorios

modeloefcaleatorios <- plm( NY.GDP.PCAP.CD ~ date, data=datos_panel , model = "random")
summary(modeloefcaleatorios)

## Oneway (individual) effect Random Effect Model 
##    (Swamy-Arora's transformation)
## 
## Call:
## plm(formula = NY.GDP.PCAP.CD ~ date, data = datos_panel, model = "random")
## 
## Balanced Panel: n = 3, T = 64, N = 192
## 
## Effects:
##                     var   std.dev share
## idiosyncratic  93420218      9665 0.375
## individual    155441504     12468 0.625
## theta: 0.9035
## 
## Residuals:
##      Min.   1st Qu.    Median   3rd Qu.      Max. 
## -24225.08  -3320.91   -892.17   5059.72  23751.53 
## 
## Coefficients:
##              Estimate Std. Error z-value  Pr(>|z|)    
## (Intercept)  1873.296   9107.904  0.2057 0.8370424    
## date1961       19.689   7891.777  0.0025 0.9980093    
## date1962       93.003   7891.777  0.0118 0.9905973    
## date1963      182.117   7891.777  0.0231 0.9815890    
## date1964      329.256   7891.777  0.0417 0.9667208    
## date1965      493.812   7891.777  0.0626 0.9501065    
## date1966      705.548   7891.777  0.0894 0.9287617    
## date1967      836.074   7891.777  0.1059 0.9156280    
## date1968     1051.287   7891.777  0.1332 0.8940250    
## date1969     1278.661   7891.777  0.1620 0.8712866    
## date1970     1483.079   7891.777  0.1879 0.8509338    
## date1971     1757.600   7891.777  0.2227 0.8237590    
## date1972     2139.145   7891.777  0.2711 0.7863449    
## date1973     2652.616   7891.777  0.3361 0.7367774    
## date1974     3306.205   7891.777  0.4189 0.6752578    
## date1975     3736.686   7891.777  0.4735 0.6358628    
## date1976     4425.604   7891.777  0.5608 0.5749430    
## date1977     4698.806   7891.777  0.5954 0.5515726    
## date1978     5234.634   7891.777  0.6633 0.5071370    
## date1979     6060.354   7891.777  0.7679 0.4425272    
## date1980     7072.576   7891.777  0.8962 0.3701483    
## date1981     8188.133   7891.777  1.0376 0.2994785    
## date1982     7987.390   7891.777  1.0121 0.3114828    
## date1983     8523.654   7891.777  1.0801 0.2801120    
## date1984     9312.706   7891.777  1.1801 0.2379796    
## date1985     9796.257   7891.777  1.2413 0.2144858    
## date1986     9909.818   7891.777  1.2557 0.2092195    
## date1987    10895.002   7891.777  1.3806 0.1674170    
## date1988    12362.836   7891.777  1.5665 0.1172207    
## date1989    13585.668   7891.777  1.7215 0.0851607 .  
## date1990    14316.347   7891.777  1.8141 0.0696648 .  
## date1991    14759.335   7891.777  1.8702 0.0614537 .  
## date1992    14990.000   7891.777  1.8994 0.0575059 .  
## date1993    15667.517   7891.777  1.9853 0.0471115 *  
## date1994    16091.651   7891.777  2.0390 0.0414460 *  
## date1995    15978.167   7891.777  2.0247 0.0429023 *  
## date1996    16773.055   7891.777  2.1254 0.0335546 *  
## date1997    17769.387   7891.777  2.2516 0.0243455 *  
## date1998    18030.354   7891.777  2.2847 0.0223303 *  
## date1999    19236.904   7891.777  2.4376 0.0147856 *  
## date2000    20835.037   7891.777  2.6401 0.0082883 ** 
## date2001    21096.198   7891.777  2.6732 0.0075134 ** 
## date2002    21538.969   7891.777  2.7293 0.0063470 ** 
## date2003    23202.118   7891.777  2.9400 0.0032817 ** 
## date2004    25366.654   7891.777  3.2143 0.0013076 ** 
## date2005    27852.977   7891.777  3.5294 0.0004166 ***
## date2006    30232.924   7891.777  3.8309 0.0001277 ***
## date2007    32408.252   7891.777  4.1066 4.016e-05 ***
## date2008    33394.731   7891.777  4.2316 2.320e-05 ***
## date2009    30291.171   7891.777  3.8383 0.0001239 ***
## date2010    33440.081   7891.777  4.2373 2.262e-05 ***
## date2011    35778.148   7891.777  4.5336 5.799e-06 ***
## date2012    36526.334   7891.777  4.6284 3.685e-06 ***
## date2013    37214.927   7891.777  4.7157 2.409e-06 ***
## date2014    37345.549   7891.777  4.7322 2.221e-06 ***
## date2015    35011.917   7891.777  4.4365 9.143e-06 ***
## date2016    34666.237   7891.777  4.3927 1.119e-05 ***
## date2017    36493.760   7891.777  4.6243 3.759e-06 ***
## date2018    38068.376   7891.777  4.8238 1.408e-06 ***
## date2019    38902.406   7891.777  4.9295 8.245e-07 ***
## date2020    37056.865   7891.777  4.6956 2.658e-06 ***
## date2021    42836.438   7891.777  5.4280 5.699e-08 ***
## date2022    46436.696   7891.777  5.8842 4.000e-09 ***
## date2023    48123.578   7891.777  6.0979 1.074e-09 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Total Sum of Squares:    5.0797e+10
## Residual Sum of Squares: 1.1958e+10
## R-Squared:      0.76459
## Adj. R-Squared: 0.64873
## Chisq: 415.742 on 63 DF, p-value: < 2.22e-16

Prueba de Hausman

phtest(modeloefcfijos, modeloefcaleatorios)

## 
##  Hausman Test
## 
## data:  NY.GDP.PCAP.CD ~ date
## chisq = 3.8736e-13, df = 63, p-value = 1
## alternative hypothesis: one model is inconsistent

Aplicación Shiny

(“https://hi9302-nahomi0yolanda-miguel0pulido.shinyapps.io/MODULO1/”)

Actividad patentes

#install.packages("WDI")
library(WDI)
#install.packages("wbstats")
library(wbstats)
#install.packages("tidyverse")
library(ggplot2)
#install.packages("plm")
library(plm)
#install.packages("readxl")
library(readxl)
#install.packages("lmtest")
library(lmtest)

## Loading required package: zoo

## 
## Attaching package: 'zoo'

## The following objects are masked from 'package:base':
## 
##     as.Date, as.Date.numeric

Importacion la base de datos

#file.choose()
patentes <- read_excel("/Users/nahomi/Desktop/GENERACIÓN DE ESCENARIOS /MODULO1/PATENT 3.xls")

Entender la base de datos

summary(patentes)

##      cusip            merger           employ            return       
##  Min.   :   800   Min.   :0.0000   Min.   :  0.085   Min.   :-73.022  
##  1st Qu.:368514   1st Qu.:0.0000   1st Qu.:  1.227   1st Qu.:  5.128  
##  Median :501116   Median :0.0000   Median :  3.842   Median :  7.585  
##  Mean   :514536   Mean   :0.0177   Mean   : 18.826   Mean   :  8.003  
##  3rd Qu.:754688   3rd Qu.:0.0000   3rd Qu.: 15.442   3rd Qu.: 10.501  
##  Max.   :878555   Max.   :1.0000   Max.   :506.531   Max.   : 48.675  
##                                    NA's   :21        NA's   :8        
##     patents         patentsg           stckpr              rnd           
##  Min.   :  0.0   Min.   :   0.00   Min.   :  0.1875   Min.   :   0.0000  
##  1st Qu.:  1.0   1st Qu.:   1.00   1st Qu.:  7.6250   1st Qu.:   0.6847  
##  Median :  3.0   Median :   4.00   Median : 16.5000   Median :   2.1456  
##  Mean   : 22.9   Mean   :  27.14   Mean   : 22.6270   Mean   :  29.3398  
##  3rd Qu.: 15.0   3rd Qu.:  19.00   3rd Qu.: 29.2500   3rd Qu.:  11.9168  
##  Max.   :906.0   Max.   :1063.00   Max.   :402.0000   Max.   :1719.3535  
##                                    NA's   :2                             
##     rndeflt             rndstck             sales               sic      
##  Min.   :   0.0000   Min.   :   0.125   Min.   :    1.22   Min.   :2000  
##  1st Qu.:   0.4788   1st Qu.:   5.152   1st Qu.:   52.99   1st Qu.:2890  
##  Median :   1.4764   Median :  13.353   Median :  174.06   Median :3531  
##  Mean   :  19.7238   Mean   : 163.823   Mean   : 1219.60   Mean   :3333  
##  3rd Qu.:   8.7527   3rd Qu.:  74.563   3rd Qu.:  728.96   3rd Qu.:3661  
##  Max.   :1000.7876   Max.   :9755.352   Max.   :44224.00   Max.   :9997  
##                      NA's   :157        NA's   :3                        
##       year     
##  Min.   :2012  
##  1st Qu.:2014  
##  Median :2016  
##  Mean   :2016  
##  3rd Qu.:2019  
##  Max.   :2021  
## 

sum(is.na(patentes))

## [1] 191

sapply(patentes, function(x) sum(is.na(x))) #NA's por variable

##    cusip   merger   employ   return  patents patentsg   stckpr      rnd 
##        0        0       21        8        0        0        2        0 
##  rndeflt  rndstck    sales      sic     year 
##        0      157        3        0        0

patentes1<- na.omit(patentes)

1. Construccion del Modelo de Datos en Panel

panel_patentes<- pdata.frame(patentes1, index = c("cusip", "year"))

2. Modelo de Efectos Fijos y Aleatorios

#Modelo de efectos fijos
modeloefcfijospatentes <- plm( patents ~ year + employ + rnd + sales, data=panel_patentes , model = "within")
summary(modeloefcfijospatentes)

## Oneway (individual) effect Within Model
## 
## Call:
## plm(formula = patents ~ year + employ + rnd + sales, data = panel_patentes, 
##     model = "within")
## 
## Unbalanced Panel: n = 215, T = 2-10, N = 2083
## 
## Residuals:
##      Min.   1st Qu.    Median   3rd Qu.      Max. 
## -489.4890   -3.3728   -1.2357    2.1915  173.8734 
## 
## Coefficients:
##             Estimate  Std. Error  t-value  Pr(>|t|)    
## year2013  7.0273e-01  2.0384e+00   0.3448   0.73032    
## year2014  2.1730e+00  2.0365e+00   1.0670   0.28610    
## year2015  3.3180e+00  2.0398e+00   1.6267   0.10397    
## year2016  2.3211e+00  2.0433e+00   1.1359   0.25613    
## year2017  3.4606e+00  2.0574e+00   1.6820   0.09273 .  
## year2018  3.4205e+00  2.0669e+00   1.6549   0.09812 .  
## year2019  4.1048e+00  2.0761e+00   1.9771   0.04817 *  
## year2020  1.0469e-01  2.0873e+00   0.0502   0.96000    
## year2021 -1.3158e+01  2.1088e+00  -6.2393 5.430e-10 ***
## employ    8.6377e-02  7.3322e-02   1.1780   0.23893    
## rnd      -1.8803e-01  1.4374e-02 -13.0818 < 2.2e-16 ***
## sales    -2.7831e-03  4.2984e-04  -6.4748 1.212e-10 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Total Sum of Squares:    1090400
## Residual Sum of Squares: 772430
## R-Squared:      0.29162
## Adj. R-Squared: 0.20537
## F-statistic: 63.6733 on 12 and 1856 DF, p-value: < 2.22e-16

#Modelo de efectos aleatorios
modeloefcaleatoriospatentes <- plm(patents ~ employ + rnd + sales, 
                                   data = panel_patentes, 
                                   model = "random")

summary(modeloefcaleatoriospatentes)

## Oneway (individual) effect Random Effect Model 
##    (Swamy-Arora's transformation)
## 
## Call:
## plm(formula = patents ~ employ + rnd + sales, data = panel_patentes, 
##     model = "random")
## 
## Unbalanced Panel: n = 215, T = 2-10, N = 2083
## 
## Effects:
##                   var std.dev share
## idiosyncratic  439.02   20.95 0.185
## individual    1935.07   43.99 0.815
## theta:
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.6808  0.8511  0.8511  0.8491  0.8511  0.8511 
## 
## Residuals:
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## -436.65   -3.75   -2.40    0.04    0.13  207.39 
## 
## Coefficients:
##                Estimate  Std. Error  z-value  Pr(>|z|)    
## (Intercept) 14.76023715  3.35560217   4.3987 1.089e-05 ***
## employ       0.99144826  0.05486048  18.0722 < 2.2e-16 ***
## rnd         -0.16626525  0.01494057 -11.1284 < 2.2e-16 ***
## sales       -0.00383260  0.00044067  -8.6971 < 2.2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Total Sum of Squares:    1308300
## Residual Sum of Squares: 1025300
## R-Squared:      0.21629
## Adj. R-Squared: 0.21516
## Chisq: 575.067 on 3 DF, p-value: < 2.22e-16

#Prueba de hausman
phtest(modeloefcfijospatentes, modeloefcaleatoriospatentes)

## 
##  Hausman Test
## 
## data:  patents ~ year + employ + rnd + sales
## chisq = 213.3, df = 3, p-value < 2.2e-16
## alternative hypothesis: one model is inconsistent

3. Pruebas de Heterocedasticidad y Autocorrelación Serial

# Pruebas de Heterocedasticidad para el modelo de efectos fijos
bptest(modeloefcfijospatentes)

## 
##  studentized Breusch-Pagan test
## 
## data:  modeloefcfijospatentes
## BP = 1472.4, df = 12, p-value < 2.2e-16

# Como el p-value <0.05, hay heterocedasticidad en los residuos (problema detectado)

# Pruebas de Heterocedasticidad para el modelo de efectos aleatorios
bptest(modeloefcaleatoriospatentes)

## 
##  studentized Breusch-Pagan test
## 
## data:  modeloefcaleatoriospatentes
## BP = 1468.8, df = 3, p-value < 2.2e-16

# Como el p-value <0.05, hay heterocedasticidad en los residuos (problema detectado)

# Prueba de autocorrelacion serial para el modelo de efectos fijos 
pwartest(modeloefcfijospatentes)

## 
##  Wooldridge's test for serial correlation in FE panels
## 
## data:  modeloefcfijospatentes
## F = 93.757, df1 = 1, df2 = 1866, p-value < 2.2e-16
## alternative hypothesis: serial correlation

# Como el p-value es < 0.05, hay autocorrelación serial en errores 

# Prueba de autocorrelacion serial para el modelo de efectos aleatorios 
pbnftest(modeloefcaleatoriospatentes)

## 
##  modified Bhargava/Franzini/Narendranathan Panel Durbin-Watson Test
## 
## data:  patents ~ employ + rnd + sales
## DW = 0.86292
## alternative hypothesis: serial correlation in idiosyncratic errors

# Como el valor es <1.5 hay autocorrelacion positiva significativa


# Correcion del modelo con errores estandar robustos
coef_corregidos <- coeftest(modeloefcfijospatentes, vcov = vcovHC(modeloefcfijospatentes, type = "HC0"))
solo_coef <- coef_corregidos[,1]

4. Generar Pronósticos y Evaluar Modelo

datos_de_prueba <- data.frame(merger=0, employ=10, return=6,
patentsg = 24, stckpr = 48, rnd = 3, rndeflt = 3, rndstck = 16, sales = 344)

length(datos_de_prueba)

## [1] 9

print(datos_de_prueba)

##   merger employ return patentsg stckpr rnd rndeflt rndstck sales
## 1      0     10      6       24     48   3       3      16   344

length(solo_coef)

## [1] 12

print(solo_coef)

##      year2013      year2014      year2015      year2016      year2017 
##   0.702733876   2.172965024   3.318045807   2.321062844   3.460617110 
##      year2018      year2019      year2020      year2021        employ 
##   3.420536926   4.104813691   0.104687590 -13.157718147   0.086376625 
##           rnd         sales 
##  -0.188032644  -0.002783103

prediccion <- sum(solo_coef*c(datos_de_prueba$merger,datos_de_prueba$employ, datos_de_prueba$return, datos_de_prueba$patentsg, datos_de_prueba$stckpr, datos_de_prueba$rnd, datos_de_prueba$rndeflt, datos_de_prueba$rndstck, datos_de_prueba$sales))


prediccion

## [1] -4240.448