#数据的量纲对估计的影响
x1<-seq(1,5,length=30)
y1<-2+3*x1+rnorm(30,0,2)        #真实的数据生成过程    系数为2和3
lm.1<-lm(y1~x1)
summary(lm.1)
            Call:
            lm(formula = y1 ~ x1)

            Residuals:
                Min      1Q  Median      3Q     Max
            -3.5420 -0.8117  0.3164  0.9730  3.1581

            Coefficients:
                        Estimate Std. Error t value Pr(>|t|)
            (Intercept)   1.6337     0.8470   1.929    0.064 .
            x1            3.1586     0.2623  12.040 1.38e-12 ***
            ---
            Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

            Residual standard error: 1.715 on 28 degrees of freedom
            Multiple R-squared: 0.8381,     Adjusted R-squared: 0.8323
            F-statistic:   145 on 1 and 28 DF,  p-value: 1.376e-12

x2<-10*x1        #同时放大十倍
y2<-10*y1
lm.2<-lm(y2~x2)
summary(lm.2)

            Call:
            lm(formula = y2 ~ x2)

            Residuals:
                Min      1Q  Median      3Q     Max
            -35.420  -8.117   3.164   9.730  31.581

            Coefficients:
                        Estimate Std. Error t value Pr(>|t|)
            (Intercept)  16.3375     8.4701   1.929    0.064 .
            x2            3.1586     0.2623  12.040 1.38e-12 ***
            ---
            Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

            Residual standard error: 17.15 on 28 degrees of freedom
            Multiple R-squared: 0.8381,     Adjusted R-squared: 0.8323
            F-statistic:   145 on 1 and 28 DF,  p-value: 1.376e-12

x3<-x2          #x放大十倍
y3<-y1
lm.3<-lm(y3~x3)
summary(lm.3)
            Call:
            lm(formula = y3 ~ x3)

            Residuals:
                Min      1Q  Median      3Q     Max
            -3.5420 -0.8117  0.3164  0.9730  3.1581

            Coefficients:
                        Estimate Std. Error t value Pr(>|t|)
            (Intercept)  1.63375    0.84701   1.929    0.064 .
            x3           0.31586    0.02623  12.040 1.38e-12 ***
            ---
            Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

            Residual standard error: 1.715 on 28 degrees of freedom
            Multiple R-squared: 0.8381,     Adjusted R-squared: 0.8323
            F-statistic:   145 on 1 and 28 DF,  p-value: 1.376e-12

    你的这个问题一定要换，一定要用人民币汇率（最好实际有效汇率，粗糙地就使用实际汇率）进行调整，总之你的经济模型的数据应该包含汇率、物价指数等等的信息。

不管怎么样，你一定要保证你模型的经济学含义是明确的，不换的话可能丧失一些信息，而且你模型的经济学含义不是很明确。在做涉及进出口问题时，我们一般把出口换算成人民币表示，然后用物价指数调整成可比序列