python 线性回归示例_线性回归示例（Julia, Python, R）

如何用最小二乘法拟合下列数据？x ：258.0, 270.0, 294.0, 320.0, 342.0, 368.0, 396.0, 446.0, 480.0, 586.0

y ：236.4, 234.4, 252.8, 298.6, 314.2, 342.2, 360.8, 368.0, 391.2, 390.8

下面采用Julia, Python, R三种科学计算语言进行处理。

julia代码如下：using DataFrames, GLM;using Plots;# Datasetsx = [258.0, 270.0, 294.0, 320.0, 342.0, 368.0, 396.0, 446.0, 480.0, 586.0];y = [236.4, 234.4, 252.8, 298.6, 314.2, 342.2, 360.8, 368.0, 391.2, 390.8];data = DataFrame(X=x, Y=y);# modelols = fit(LinearModel, @formula(Y ~ X), data); # ols = lm(@formula(Y ~ X), data)ypredict = round.(predict(ols), digits=5);println(ols);println(coef(ols));# plotscatter(x, y, marker=:c , c=:blue);plot!(x, ypredict, ls=:solid, c=:red, frame>xlabel!("x");ylabel!("y");savefig("example8_julia");

结果如下：Y ~ 1 + X

Coefficients:

──────────────────────────────────────

Estimate  Std. Error  t value  Pr(>|t|)  Lower 95%   Upper 95%

──────────────────────────────────────

(Intercept)  115.786     33.3936     3.4673     0.0085   38.7797   192.791

X              0.540304   0.0859298  6.28774    0.0002    0.34215    0.738459

──────────────────────────────────────

[115.78555030272875, 0.5403043874927427]

=

python代码如下：# encoding:utf8import numpy as npimport matplotlib.pyplot as pltfrom sklearn import linear_modelfrom sklearn.metrics import mean_squared_error, r2_score# Datasetsx = np.array([258.0, 270.0, 294.0, 320.0, 342.0, 368.0, 396.0, 446.0, 480.0, 586.0])[:, np.newaxis]y = np.array([236.4, 234.4, 252.8, 298.6, 314.2, 342.2, 360.8, 368.0, 391.2, 390.8])# modelreg = linear_model.LinearRegression()reg.fit(x, y)ypredict = reg.predict(x)print("Coefficients: \n", reg.coef_) # The coefficientsprint("Mean squared error: %.2f"% mean_squared_error(y, ypredict)) # The mean squared errorprint("Variance score: %.2f" % r2_score(y, ypredict)) # Explained variance score: 1 is perfect prediction# plotplt.figure(figsize=(10, 6))plt.plot(x, y, "o")plt.plot(x, ypredict, "-")plt.xlabel("x", fontsize = 20)plt.ylabel("y", fontsize = 20)plt.savefig("example8_python.png")plt.show()

结果如下：Coefficients: [0.54030439]

Mean squared error: 569.78

Variance score: 0.83

=

R代码如下：# Datasetsx

结果如下：> print(summary(lm(y~x)))

Call:

lm(formula = y ~ x)

Residuals:

Min     1Q Median     3Q    Max

-41.60 -21.07  10.58  15.46  31.05

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 115.78555   33.39361   3.467 0.008476 **

x             0.54030    0.08593   6.288 0.000236 ***

---

Signif. codes:  0 "***" 0.001 "**" 0.01 "*" 0.05 "." 0.1 " " 1

Residual standard error: 26.69 on 8 degrees of freedom

Multiple R-squared:  0.8317,    Adjusted R-squared:  0.8107

F-statistic: 39.54 on 1 and 8 DF,  p-value: 0.0002359

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

简称：NLF；

爱好：Coding，数据分析，散步，动漫迷，冥想

；

更刊说明：受动漫启发，推行一周一刊制度，每刊不限一条内容，每周日定期推刊，偶尔也会休刊

，多多关照！

刊物内容：主推技术流(涉及基于Julia、Python、R、Matlab、C\C++等语言的科普文)，学科流(OpenFOAM等流体力学求解器知识、生态水文学习中)，统计学；

近期规划：完成硕士毕业论文，备考生态水文学方向的博士。

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~