Mini project 1

Question 1

What is the median of the first column using the car dataset

data(cars)
median(cars$speed)

## [1] 15

Question 2

What is the maximum of daily close price for BTC in the data?

install.packages(“jsonlite”)

library(jsonlite)

url <- "https://min-api.cryptocompare.com/data/v2/histoday?fsym=BTC&tsym=USD&limit=10"

btc_data_raw <- fromJSON(url)

# Extract the actual data frame 
btc_data <- btc_data_raw$Data$Data

# View the first few rows 
head(btc_data)

##         time     high      low     open volumefrom   volumeto    close
## 1 1771632000 68686.92 67528.59 67997.02   11686.08  796953224 67965.05
## 2 1771718400 68233.91 67170.64 67965.05   11615.92  786667937 67626.88
## 3 1771804800 67663.82 63879.72 67626.88   42373.83 2766008869 64642.48
## 4 1771891200 64996.35 62551.85 64642.48   45607.54 2906212703 64069.84
## 5 1771977600 70005.48 63922.45 64069.84   55809.16 3753044531 67998.99
## 6 1772064000 68867.59 66508.93 67998.99   44886.39 3038274772 67498.16
##   conversionType conversionSymbol
## 1         direct                 
## 2         direct                 
## 3         direct                 
## 4         direct                 
## 5         direct                 
## 6         direct

# Calculate maximum daily close price
max_close <- max(btc_data$close, na.rm = TRUE)

# Print result
max_close

## [1] 68915.1

Question 3

Title: Do People use Social Media More During Winter?

Research Questions

1. Is search interest for Instagram higher in winter than summer?

2. Is there a seasonal pattern in social media search behavior?

3. Do colder months show consistently higher engagement levels?

Download the data manually from Google Trends using the keyword “Instagram”. Import the data as a CSV file.

trend_data <- read.csv("multiTimeline.csv", skip = 1)

head(trend_data)

##         Week Instagram...United.States.
## 1 2021-02-28                         64
## 2 2021-03-07                         59
## 3 2021-03-14                         60
## 4 2021-03-21                         60
## 5 2021-03-28                         62
## 6 2021-04-04                         62

Rename the columns so it’s easier to read

colnames(trend_data) <- c("date", "hits")

trend_data$date <- as.Date(trend_data$date)

trend_data$hits <- as.numeric(trend_data$hits)

Create a month variable/column in the data

trend_data$month <- format(trend_data$date, "%m")

Using those months, categorize them into seasons.

trend_data$season <- ifelse(trend_data$month %in% c("12","01","02"), "Winter",
                     ifelse(trend_data$month %in% c("03","04","05"), "Spring",
                     ifelse(trend_data$month %in% c("06","07","08"), "Summer",
                            "Fall")))

Compare the seasons using the mean

season_avg <- aggregate(hits ~ season, data = trend_data, mean)

season_avg

##   season     hits
## 1   Fall 54.29231
## 2 Spring 55.07576
## 3 Summer 56.81818
## 4 Winter 54.87692

The data shows the mean is the highest for summer, then spring, then winter, then fall.

Create a graph to show the difference visually.

bp <- barplot(season_avg$hits,
              names.arg = season_avg$season,
              main = "Average Instagram Search Interest by Season",
              xlab = "Season",
              ylab = "Average Search Interest",
              col = c("lightblue","lightgreen","orange","tan"),
              ylim = c(min(season_avg$hits) - 2,
                       max(season_avg$hits) + 2))

text(bp,
     season_avg$hits,
     round(season_avg$hits, 1),
     pos = 3)

Since the y values are so close together, I zoomed in so you can see the difference in the seasons. As shown, summer is the highest with fall in the lowest.