gudimetlatejaswi
commented
7 months ago assignment 2
install.packages("tidyverse") #> Installing package into 'C:/Users/999te/AppData/Local/R/win-library/4.3' #> (as 'lib' is unspecified) #> package 'tidyverse' successfully unpacked and MD5 sums checked #> #> The downloaded binary packages are in #> C:\Users\999te\AppData\Local\Temp\RtmpSOufbc\downloaded_packages library(tidyverse) library(prophet) #> Loading required package: Rcpp #> Loading required package: rlang #> #> Attaching package: 'rlang' #> The following objects are masked from 'package:purrr': #> #> %@%, flatten, flatten_chr, flatten_dbl, flatten_int, flatten_lgl, #> flatten_raw, invoke, splice # Read data from CSV file df <- read.csv("C:/Users/999te/Downloads/ICNSA.csv") # Rename columns for compatibility with Prophet df <- df %>% rename(ds = DATE, y = ICNSA) # Fit a prophet model m_prophet <- prophet(df) #> Disabling weekly seasonality. Run prophet with weekly.seasonality=TRUE to override this. #> Disabling daily seasonality. Run prophet with daily.seasonality=TRUE to override this. future_prophet <- make_future_dataframe(m_prophet, periods = 1) forecast_prophet <- predict(m_prophet, future_prophet) # Display prophet forecast cat("Prophet Forecast:\n") #> Prophet Forecast: tail(forecast_prophet[c('ds', 'yhat', 'yhat_lower', 'yhat_upper')]) #> ds yhat yhat_lower yhat_upper #> 2974 2023-12-30 544996.4 223329.3 844870.0 #> 2975 2024-01-06 594020.4 289690.8 900591.5 #> 2976 2024-01-13 593485.3 289755.4 881218.5 #> 2977 2024-01-20 538122.9 226946.8 826686.9 #> 2978 2024-01-27 465529.1 181721.1 782285.9 #> 2979 2024-01-28 456659.9 138969.4 765446.7 future_predicted_value <- forecast_prophet$yhat[nrow(forecast_prophet)] last_forecast <- tail(forecast_prophet, 1) # Extracting the yhat value future_predicted_value <- last_forecast$yhat[1] # Specify the released forecast value released_forecast_value <- 261029 # Calculate the difference (yhat - released forecast value) predicted_forecast <- future_predicted_value - released_forecast_value cat("\nFuture Predicted Value (yhat - Released Forecast Value):", predicted_forecast, "\n") #> #> Future Predicted Value (yhat - Released Forecast Value): 195630.9 # Add a lag column to the data frame if 'y' column exists if ("y" %in% colnames(df)) { df$y_lag <- lag(df$y) # Remove rows with NAs introduced by lag df <- na.omit(df) } else { cat("Column 'y' not found in the dataframe.\n") } # Plot observed and predicted values from both Prophet and linear regression df_lm <- df %>% mutate(y_lm = predict(lm(y ~ y_lag, data = df), newdata = df)) library(ggplot2) library(dplyr) # Assuming df_lm is your dataframe and it has 'ds' for date, 'y' for actual and 'y_lm' for predicted # Convert 'ds' from POSIXct/POSIXlt to Date if it's not already df_lm$ds <- as.Date(df_lm$ds) # Plotting with ggplot2 ggplot(df_lm, aes(x = ds)) + geom_line(aes(y = y, color = "Actual"), linewidth = 1, group = 1) + # Add group = 1 geom_line(aes(y = y_lm, color = "Predicted"), linetype = "dashed", linewidth = 1, group = 1) + # Add group = 1 labs(title = "Actual vs. Predicted Values", y = "ICNSA", x = "Date", color = "Values") + scale_color_manual(values = c("Actual" = "black", "Predicted" = "red")) + theme_minimal()
Created on 2024-02-11 with reprex v2.1.0
- Predicted future forecast value is 195630.9
Assignment1