Example 2: The nearby data shows how the weight of a newborn at various numbers of days after birth.

1 Obtain the regression formula and correlation coefficient produced by linear regression of the data.

2 Obtain the regression formula and correlation coefficient produced by logarithmic regression of the data.

3 Predict the weight 350 days after birth based on the regression formula that best fits the trend of the data in accordance with the regression results.

Operation Procedure

Select the REG Mode: N4(REG)

Select FreqOff for the statistical frequency setting: 1N(SETUP)c5(STAT)2(FreqOff)

Number

Weight

of Days

(g)

20

3150

50

4800

80

6420

110

7310

140

7940

170

8690

200

8800

230

9130

260

9270

290

9310

320

9390

Input the number of day data into the X-column: 20E50E80E110E140E170E 200E230E260E290E320E

Input the weight data into the Y-column: ce3150E4800E6420E7310E 7940E8690E8800E9130E 9270E9310E9390E

(1) Linear RegressionDisplay the linear regression calculation result screen:z6(RESULT)2(Reg)1(Line)(2) Distribution Logarithmic RegressionDisplay the logarithmic regression calculation result screen:Jz6(RESULT)2(Reg)3(Log)(3) Weight Prediction

The absolute value of correlation coefficient r is close to 1, so logarithmic regression is used for its calculation.

Obtain n when x = 350:Jz1(/COMP)350z7(STAT) 2(VAR)ccc7(n)E

E-85