R 软件 2025 最新下载指南：包生态入门与可视化工具实操教程

library(ggplot2)
ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point(aes(color = cyl)) +  # 按气缸数（cyl）区分颜色
  labs(title = "马力与燃油效率关系", x = "马力", y = "燃油效率（mpg）") +
  theme_minimal()

ggplot(data = mtcars, aes(x = factor(cyl), fill = factor(cyl))) +
  geom_bar() +
  labs(title = "不同气缸数汽车数量统计", x = "气缸数", y = "数量") +
  scale_fill_manual(values = c("#FF6B6B", "#4ECDC4", "#C7F464")) +  # 自定义颜色
  theme_minimal()

ggplot(data = mtcars, aes(x = hp, y = mpg)) +
  geom_point(aes(color = cyl)) +
  geom_smooth(method = "lm", se = FALSE) +  # 添加线性回归线
  labs(title = "马力与燃油效率关系及回归线", x = "马力", y = "燃油效率（mpg）") +
  theme_minimal()

library(plotly)
data <- read.csv("stock_data.csv")  # 假设数据已导入
fig <- plot_ly(data, x = ~date, y = ~close, type = "scatter", mode = "lines")
fig <- fig %>% layout(title = "股票收盘价走势", xaxis = list(title = "日期"), yaxis = list(title = "收盘价"))
fig

data <- data.frame(
  x = rnorm(),
  y = rnorm(),
  z = rnorm(),
  color = sample(c("A", "B", "C"), , replace = TRUE)
)
fig <- plot_ly(data, x = ~x, y = ~y, z = ~z, color = ~color, type = "scatter3d", mode = "markers")
fig <- fig %>% layout(scene = list(xaxis = list(title = "X"), yaxis = list(title = "Y"), zaxis = list(title = "Z")))
fig

data <- data.frame(
  year = rep(:, each = ),
  value = rnorm(),
  group = rep(letters[:], )
)
fig <- plot_ly(data, x = ~group, y = ~value, color = ~group, type = "scatter", mode = "markers")
fig <- fig %>% animation_opts(frame = ~year)
fig

data <- read.csv("credit_data.csv")
# 处理缺失值
data <- na.omit(data)
# 转换分类变量
data$default <- factor(data$default, levels = c(, ), labels = c("未违约", "违约"))

# 违约率统计
table(data$default)
# 绘制违约率饼图
ggplot(data, aes(x = "", fill = default)) +
  geom_bar(width = ) +
  coord_polar("y", start = ) +
  labs(title = "信用卡违约率分布") +
  theme_void()

# 绘制年龄与违约率的箱线图
ggplot(data, aes(x = default, y = age)) +
  geom_boxplot() +
  labs(title = "年龄与违约率的关系") +
  theme_minimal()

# 绘制收入与违约率的散点图
ggplot(data, aes(x = income, y = default)) +
  geom_jitter() +
  labs(title = "收入与违约率的关系") +
  theme_minimal()

library(caret)
# 划分训练集和测试集
set.seed()
trainIndex <- createDataPartition(data$default, p = 0.8, list = FALSE)
trainData <- data[trainIndex, ]
testData <- data[-trainIndex, ]

# 建立逻辑回归模型
model <- glm(default ~ age + income + credit_limit, data = trainData, family = "binomial")
summary(model)

# 模型预测与评估
predictions <- predict(model, testData, type = "response")
confusionMatrix(data = factor(ifelse(predictions > 0.5, "违约", "未违约")), reference = testData$default)