A Confidence Interval for the Slope · 回归斜率的置信区间
The LINER conditions
- Inference about a slope needs five conditions — remember LINER:
- Linear (the true relationship is linear) · Independent observations.
- Normal (responses vary normally around the line) · Equal SD (constant scatter across $x$).
- Random (data from a random sample or experiment). Residual plots help check L, N, E.
LINER 条件
- 关于斜率的推断需要五个条件——记住 LINER:
- Linear(真实关系是线性的)· Independent(观测独立)。
- Normal(响应在线周围正态变动)· Equal SD(跨 $x$ 散布恒定)。
- Random(数据来自随机样本或实验)。残差图有助于检查 L、N、E。
Read the computer output
- Slope inference almost always starts from computer output.
- Find the row for the explanatory variable: its estimate is the slope $b$.
- The next column is the standard error $SE_b$.
- Output also reports a $t$ statistic and p-value — but first we build an interval.
读计算机输出
- 斜率推断几乎总是从计算机输出开始。
- 找到解释变量那一行:它的估计值就是斜率 $b$。
- 下一列是标准误 $SE_b$。
- 输出也报告一个 $t$ 统计量和 p 值——但我们先建区间。
Build the t-interval
- The slope interval is estimate $\pm$ margin of error:
-
$$b \pm t^{*}\, SE_b$$
- Use the $t$-distribution with $df = n - 2$ (two estimated: slope and intercept).
- $t^{*}$ comes from that $df$ at your confidence level.
构建 t 区间
- 斜率区间是估计 $\pm$ 误差幅度:
-
$$b \pm t^{*}\, SE_b$$
- 使用 $df = n - 2$ 的 $t$ 分布(估计了两个:斜率和截距)。
- $t^{*}$ 来自那个 $df$、在你的置信水平下。
Interpret in context
- "We're $95\%$ confident the true slope $\beta$ is between the endpoints."
- Say what the slope means: the predicted change in $y$ per one-unit increase in $x$.
- Name the real variables and units.
- The interval is a range of plausible values for $\beta$.
结合语境解读
- “我们有 $95\%$ 把握,真实斜率 $\beta$ 在两个端点之间。”
- 说清斜率的含义:$x$ 每增加一个单位时 $y$ 的预测变化量。
- 点明真实变量和单位。
- 区间是 $\beta$ 的一个合理取值范围。
Slope inference uses $df = n - 2$, not $n - 1$. Two parameters are estimated (slope and intercept), so you lose two degrees of freedom. Read $b$ and $SE_b$ straight from the output — don't recompute them — and remember the interval is $b \pm t^{*}SE_b$, the same estimate-$\pm$-margin form as every other $t$-interval.
**斜率推断用 $df = n - 2$,而非 $n - 1$。**估计了两个参数(斜率和截距),所以损失两个自由度。直接从输出读 $b$ 和 $SE_b$——不要重新计算它们——并记住区间是 $b \pm t^{*}SE_b$,与其他每个 $t$ 区间相同的估计-$\pm$-幅度形式。
Output: slope $b = 4.2$, $SE_b = 1.0$, $n = 20$. Build a $95\%$ interval. ($df = 18$, $t^{*} \approx 2.10$.)
- Margin of error: $2.10 \times 1.0 = 2.10$.
- Interval: $4.2 \pm 2.10 = (2.1,\ 6.3)$.
- Interpret: we're $95\%$ confident each extra hour predicts between $2.1$ and $6.3$ more points.
输出:斜率 $b = 4.2$,$SE_b = 1.0$,$n = 20$。建一个 $95\%$ 区间。($df = 18$,$t^{*} \approx 2.10$。)
- 误差幅度:$2.10 \times 1.0 = 2.10$。
- 区间:$4.2 \pm 2.10 = (2.1,\ 6.3)$。
- **解读:**我们有 $95\%$ 把握,每多一个小时预测多得 $2.1$ 到 $6.3$ 分。
After the LINER conditions, read $b$ and $SE_b$ from the output and build a slope $t$-interval $b \pm t^{*}\,SE_b$ using $df = n - 2$. Interpret it in context — a range of plausible values for the true slope $\beta$ (the predicted change in $y$ per unit $x$).
在 LINER 条件之后,从输出读 $b$ 和 $SE_b$,用 $df = n - 2$ 建斜率 $t$ 区间 $b \pm t^{*}\,SE_b$。结合语境解读它——真实斜率 $\beta$($x$ 每单位 $y$ 的预测变化量)的一个合理取值范围。
The fitted slope and its uncertainty · 拟合斜率及其不确定性
The interval b ± t*·SE_b gives plausible values for the true slope. · 区间 b ± t*·SE_b 给出真实斜率的合理取值。
For slope inference with n = 20, what are the degrees of freedom (n − 2)? · 对 n = 20 的斜率推断,自由度(n − 2)是多少?
df = n − 2 = 18 (slope and intercept estimated). · df = n − 2 = 18(估计了斜率和截距)。
b = 4.2, SE_b = 1.0, t* = 2.10. Find the margin of error t*·SE_b. · b = 4.2,SE_b = 1.0,t* = 2.10。求误差幅度 t*·SE_b。
2.10 × 1.0 = 2.10. · 2.10 × 1.0 = 2.10。
In computer regression output, the slope b is found in the column labeled... · 在计算机回归输出中,斜率 b 在标记为……的列。
The explanatory variable's estimate is the slope. · 解释变量的估计值就是斜率。
Slope inference uses n − 1 degrees of freedom. · 斜率推断使用 n − 1 自由度。
It's n − 2 — two parameters (slope and intercept) are estimated. · 是 n − 2——估计了两个参数(斜率和截距)。
The mnemonic for the five slope-inference conditions is ___ (five letters). · 五个斜率推断条件的助记词是 ___(五个字母)。
Linear, Independent, Normal, Equal SD, Random. · 线性、独立、正态、等标准差、随机。