使用R Plotly绘制二次回归线：新手实操技术咨询

Hey there! Since you're already familiar with fitting the second-order polynomial models using lm(), let's put those fits to work with Plotly to create clear, interactive plots for both your variable pairs. Here's a complete, step-by-step solution tailored to your data:

Step 1: Load Required Packages & Prepare Your Data #

First, make sure you have Plotly installed (run install.packages("plotly") if you haven't already), then load it along with basic data-handling tools:

# Install package if missing
# install.packages("plotly")
library(plotly)

# Define your data and wrap it in a data frame (simplifies plotting)
price <- c(995, 675, 690, 600, 612, 700, 589, 532, 448, 512, 537, 560)
score <- c(89, 91, 88, 89, 91, 91, 89, 93, 83, 91, 91, 90)
year <- c(2005:2016)

df <- data.frame(year, score, price)

Step 2: Fit the Second-Order Polynomial Models #

You mentioned you already have qfit1 working, but let's include both fits explicitly. We use poly(..., raw = TRUE) to work with intuitive raw polynomial terms instead of orthogonal ones:

# Price vs. Year quadratic fit
qfit1 <- lm(price ~ poly(year, 2, raw = TRUE), data = df)

# Price vs. Score quadratic fit
qfit2 <- lm(price ~ poly(score, 2, raw = TRUE), data = df)

Step 3: Generate Smooth Prediction Sequences #

To create a clean, smooth regression line (not just connecting original data points), we'll generate dense sequences of predictor values and use our models to predict corresponding prices:

# Smooth sequence for year (covers full range of your data)
year_seq <- seq(min(df$year), max(df$year), length.out = 100)
pred_year <- predict(qfit1, newdata = data.frame(year = year_seq))

# Smooth sequence for score
score_seq <- seq(min(df$score), max(df$score), length.out = 100)
pred_score <- predict(qfit2, newdata = data.frame(score = score_seq))

Step 4: Create Interactive Plotly Plots #

We'll build two separate interactive plots, plus an option to combine them into side-by-side subplots:

Option 1: Separate Plots

First, the Price vs. Year plot:

plot_year <- plot_ly(df, x = ~year, y = ~price, 
                     type = "scatter", mode = "markers", 
                     name = "Raw Data", marker = list(color = "darkgray")) %>%
  add_trace(x = ~year_seq, y = ~pred_year,
            type = "scatter", mode = "lines",
            name = "Quadratic Fit", line = list(color = "#e74c3c", width = 2)) %>%
  layout(title = "Price vs. Year: Second-Order Polynomial Fit",
         xaxis = list(title = "Year"),
         yaxis = list(title = "Price"),
         showlegend = TRUE)

# Display the plot
plot_year

Then the Price vs. Score plot:

plot_score <- plot_ly(df, x = ~score, y = ~price, 
                      type = "scatter", mode = "markers", 
                      name = "Raw Data", marker = list(color = "darkgray")) %>%
  add_trace(x = ~score_seq, y = ~pred_score,
            type = "scatter", mode = "lines",
            name = "Quadratic Fit", line = list(color = "#3498db", width = 2)) %>%
  layout(title = "Price vs. Score: Second-Order Polynomial Fit",
         xaxis = list(title = "Score"),
         yaxis = list(title = "Price"),
         showlegend = TRUE)

# Display the plot
plot_score

Option 2: Combined Subplots

If you want to compare both fits side-by-side (or stacked), use subplot():

subplot(plot_year, plot_score, nrows = 2, shareY = TRUE) %>%
  layout(title = list(text = "Quadratic Polynomial Fits: Price vs. Year & Score", x = 0.5),
         margin = list(t = 60))

A Quick Tip #

All plots are interactive—you can hover over points to see exact values, zoom, pan, and toggle the legend to show/hide raw data or the fit line. Feel free to tweak colors, line widths, or titles to match your preferences!

内容的提问来源于stack exchange，提问作者Giannis K