Module 10. Time Series and Visualization

 





Visualization plays a big role in time series analysis because it helps us actually see how data changes over time instead of just looking at numbers. By plotting data, we can quickly spot patterns, trends, or seasonal behaviors that might not be obvious from raw values alone. For example, a line chart can show whether unemployment rates rise or fall during certain periods, helping us understand the bigger picture. Visuals also make it easier to detect sudden changes or outliers that could affect the analysis. Overall, visualization turns complex time-based data into something more intuitive and meaningful, making it a key first step in analyzing and interpreting time series data.


R code for future reference:

#Hot Dog eating contest
library(readr)
library(ggplot2)

# Load the dataset
hotdogs <- read_csv("http://datasets.flowingdata.com/hot-dog-contest-winners.csv")
#view
head(hotdogs)

# Set color for bars: dark red if new record, grey otherwise
colors <- ifelse(hotdogs$`New record` == 1, "darkred", "grey")
heights <- as.numeric(hotdogs$`Dogs eaten`)

# Create the bar plot
barplot(
  heights,
  names.arg = hotdogs$Year,
  col = colors,
  border = NA,
  main = "Nathan's Hot Dog Eating Contest Results, 1980–2010",
  xlab = "Year",
  ylab = "Hot dogs and buns (HDBs) eaten"
)

#ggplot2
ggplot(hotdogs, aes(x = Year, y = `Dogs eaten`, fill = factor(`New record`))) +
  geom_col() +  # same as geom_bar(stat = "identity")
  labs(
    title = "Nathan's Hot Dog Eating Contest Results, 1980-2010",
    x = "Year",
    y = "Hot dogs and buns (HDBs) eaten",
    fill = "New Record"
  )

#Stacked bar plot using R-base graphics
hotdogs_2000 <- subset(hotdogs, Year >= 2000 & Year <= 2010)
# Convert to matrix
hotdog_places <- as.matrix(hotdog_places)
colnames(hotdog_places) <- as.character(2000:2010)

#create plot
barplot(
  hotdog_places,
  border = NA,
  main = "Hot Dog Eating Contest Results, 1980–2010",
  xlab = "Year",
  ylab = "Hot dogs and buns (HDBs) eaten",
  col = gray.colors(3)
)

#economics data
library(ggplot2)
# Load the built-in dataset
head(economics)

# Extract the year from the date column
year <- function(x) as.POSIXlt(x)$year + 1900
economics$year <- year(economics$date)
head(economics)

# Base R plot equivalent of qplot(date, unemploy / pop, data = economics, geom = "line")
plot(economics$date, economics$unemploy / economics$pop,
     type = "l",                      
     main = "Unemployment as a Fraction of Population (1967–2015)",
     xlab = "Date",
     ylab = "unemploy/pop")

#unemployment graph
# Load data
data(economics)

# Create the year variable
year <- function(x) as.POSIXlt(x)$year + 1900
economics$year <- year(economics$date)

# Define x and y
x <- economics$unemploy / economics$pop
y <- economics$uempmed

# Create a color gradient for the years
pal <- colorRampPalette(c("#08306b", "#2171b5", "#6baed6", "#bdd7e7"))
n <- 100
brks <- seq(min(economics$year), max(economics$year), length.out = n)
col_index <- cut(economics$year, breaks = brks, include.lowest = TRUE, labels = FALSE)
cols <- pal(n)[col_index]

plot(x, y, type = "n",
     xlab = "unemploy/pop",
     ylab = "uempmed",
     main = "Median Unemployment vs. Unemployment Rate (colored by Year)")

# Draw colored path and points
for (i in 1:(length(x) - 1)) {
  segments(x[i], y[i], x[i + 1], y[i + 1], col = cols[i], lwd = 2)
}
points(x, y, pch = 16, col = cols)

# Add a simple legend for years
legend("topleft",
       legend = c(1970, 1980, 1990, 2000, 2010),
       title = "year",
       pch = 16,
       col = pal(5),
       bty = "n")



Comments

Post a Comment

Popular posts from this blog

Final Project Diamond Database

Image
 Problem Description:  I have always been interested in diamonds—their value, how they are traded, and the global network behind them. Growing up in India, where diamond buying is culturally significant and cities like Surat serve as major worldwide centers for diamond cutting and polishing, I was frequently exposed to discussions about diamond quality, pricing, and craftsmanship. I also enjoy watching documentaries about diamond grading and international trading, which deepened my curiosity about how specific characteristics influence market value. Although my personal experience is rooted in the Indian diamond market, the dataset used in this project—the well-known diamonds dataset—reflects U.S. retail pricing. This creates an interesting opportunity to compare my personal understanding with patterns observed in a different but well-established market. Importantly, many of the attributes used to evaluate diamonds—carat weight, cut quality, clarity, and color—are universal gr...
Read more

Module 6. basic data visualization using R

Image
I created a boxplot using the mtcars dataset to compare miles per gallon (MPG) among vehicles with different numbers of cylinders. A boxplot is useful for showing the distribution and variability of data across categories, highlighting the median, quartiles, and outliers. I chose this chart because it provides a clear summary of how fuel efficiency changes as engine size (cylinders) increases, without too much intricate detail. To add, the visualization did reveal clear differences between the groups. Cars with four cylinders had noticeably higher MPG values, showing that they are generally more fuel-efficient. In contrast, vehicles with six or eight cylinders had lower MPG values, suggesting that as engine size increases, fuel efficiency tends to decrease. This pattern shows a strong inverse relationship between the number of cylinders and miles per gallon. In terms of deviations in the data, the chart mostly confirmed the expected trend that larger engines consume more fuel. Howe...
Read more

Module 5. Assignment: Create Your Own Visualizations assignment Plotly vs. Datawrapper

Image
The dataset can be seen below: Ranking Chart using an ordered bar chart: Part-to-Whole Chart using a donut chart: Why you chose this dataset. I chose this dataset because it has two straightforward but meaningful variables: Time on the x-axis and Average Position on the y-axis. This setup is applicable in many real-world contexts. For example, “Time” could represent days, months, or trials in an experiment, while “Average Position” could represent a performance ranking, a score, or some measure of quality. Many real datasets look like this: something is being tracked over time to see whether it improves or declines. Practicing with this data can assist in getting used to analyzing patterns that are very relevant in fields like business (tracking customer satisfaction), sports (tracking performance stats), or technology (tracking algorithm accuracy).  The story or insights your visualizations tell. The ordered bar chart revealed that as Time increased, the Average Position steadily ...
Read more