Introduction

This report presents an analysis of the tornado dataset focusing on various aspects including exploratory data analysis, visualization, and statistical modeling.

Exploratory Data Analysis

Summary Statistics

Tornado Dataset Summary Statistics

This report displays summary statistics of the tornado dataset.

# Load or define dataset_al_tornado
dataset_al_tornado <- read.csv("C:\\Users\\Ross\\Documents\\RDatasets\\dataset_al_tornado.csv")

# Display summary statistics of the dataset
summary(dataset_al_tornado)
##  Tornado...in.Year      Year          Month             Day       
##  Min.   :  1.00    Min.   :1794   Min.   : 1.000   Min.   : 1.00  
##  1st Qu.:  5.00    1st Qu.:1961   1st Qu.: 3.000   1st Qu.: 8.00  
##  Median : 14.00    Median :1991   Median : 4.000   Median :17.00  
##  Mean   : 21.67    Mean   :1977   Mean   : 5.545   Mean   :16.61  
##  3rd Qu.: 30.00    3rd Qu.:2008   3rd Qu.: 9.000   3rd Qu.:24.00  
##  Max.   :145.00    Max.   :2018   Max.   :12.000   Max.   :31.00  
##                                   NA's   :11       NA's   :16     
##   Time..CST.           County          Damage.Scale      
##  Length:2663        Length:2663        Length:2663       
##  Class :character   Class :character   Class :character  
##  Mode  :character   Mode  :character   Mode  :character  
##                                                          
##                                                          
##                                                          
##                                                          
##  Maximum.Path.Width..Yards. Path.Length..Miles.  Fatalities       
##  Min.   :   0.0             Length:2663         Length:2663       
##  1st Qu.:  30.0             Class :character    Class :character  
##  Median : 100.0             Mode  :character    Mode  :character  
##  Mean   : 176.3                                                   
##  3rd Qu.: 200.0                                                   
##  Max.   :2600.0                                                   
##  NA's   :136                                                      
##    Injuries           Location            X          
##  Length:2663        Length:2663        Mode:logical  
##  Class :character   Class :character   NA's:2663     
##  Mode  :character   Mode  :character                 
##                                                      
##                                                      
##                                                      
##

Time series plot for tornado frequency by year

library(ggplot2)

Aggregate tornado counts per year

tornado_counts <- aggregate(Tornado...in.Year ~ Year, data = dataset_al_tornado, FUN = length)

Seasonal analysis - creating a bar plot of tornado occurrences by month

ggplot(dataset_al_tornado, aes(x = factor(Month), fill = Damage.Scale)) +
  geom_bar() +
  labs(title = "Tornado Occurrences by Month", x = "Month", y = "Tornado Count", fill = "Damage Scale") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

# Summary statistics of the Damage Scale

summary(dataset_al_tornado$Damage.Scale)
##    Length     Class      Mode 
##      2663 character character

Create a time series plot

ggplot(tornado_counts, aes(x = Year, y = Tornado...in.Year)) +
  geom_line() +
  labs(title = "Tornado Frequency Over Years", x = "Year", y = "Tornado Count")

Bar chart for total tornado count based on damage scale

damage_counts <- aggregate(Tornado...in.Year ~ Damage.Scale, data = dataset_al_tornado, FUN = length)

Plotting the bar chart

ggplot(damage_counts, aes(x = Damage.Scale, y = Tornado...in.Year)) +
  geom_bar(stat = "identity") +
  labs(title = "Tornado Count by Damage Scale", x = "Damage Scale", y = "Tornado Count")

Perform a simple linear regression

model <- lm(Tornado...in.Year ~ Month + Damage.Scale, data = dataset_al_tornado)

# Summary of the regression model
summary(model)
## 
## Call:
## lm(formula = Tornado...in.Year ~ Month + Damage.Scale, data = dataset_al_tornado)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -35.278 -11.462  -4.429   4.825 108.193 
## 
## Coefficients:
##                Estimate Std. Error t value Pr(>|t|)    
## (Intercept)     12.4202     1.0861  11.435  < 2e-16 ***
## Month            2.6887     0.1211  22.198  < 2e-16 ***
## Damage.ScaleF1  -1.9205     1.0509  -1.827   0.0678 .  
## Damage.ScaleF2 -12.0571     1.1007 -10.954  < 2e-16 ***
## Damage.ScaleF3 -10.5127     1.6224  -6.480  1.1e-10 ***
## Damage.ScaleF4  -3.7180     2.6674  -1.394   0.1635    
## Damage.ScaleF5  12.4109     7.2960   1.701   0.0891 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 20.5 on 2523 degrees of freedom
##   (133 observations deleted due to missingness)
## Multiple R-squared:  0.2266, Adjusted R-squared:  0.2248 
## F-statistic: 123.2 on 6 and 2523 DF,  p-value: < 2.2e-16

Summary: This report includes sections for summary statistics, time series plots, data visualizations, regression analysis, and a conclusion based on the tornado dataset analysis.

Dataset Analysis Summary

  1. Data Exploration Loading the Dataset: Initially, we loaded the tornado dataset into RStudio for analysis. Exploring Dataset Structure: Checked the dataset’s structure using str() to understand variable types and dimensions. Summary Statistics: Utilized summary() to get an overview of the dataset’s numeric variables.
  2. Visualization Time Series Plot: Created a time series plot to visualize tornado frequency over the years. Bar Chart: Generated a bar chart to display tornado counts based on damage scale. Seasonal Analysis: Visualized tornado occurrences by month using a bar plot to identify any seasonal patterns.
  3. Statistical Analysis Linear Regression: Conducted a simple linear regression to predict tornado occurrences based on month and damage scale. Cleaning Data: Removed rows with missing values in the ‘Damage.Scale’ column for the regression analysis.
  4. RMarkdown Setup RMarkdown Report: Started creating an RMarkdown report incorporating exploratory analysis, visualization, and regression findings. Troubleshooting: Debugged issues with code chunks within the RMarkdown file to ensure a smooth knitting process.
  5. RMarkdown File Generation Generated RMarkdown Document: Provided a template for an RMarkdown file, including code chunks for summary statistics, which can be adapted with the actual dataset.