Bias Correction in Logarithmic Regression and Comparison with Weighted Regression for Non-linear Models
-
1.
Research Institute of Forest Resource Information Techniques, Chinese Academy of Forestry, Beijing 100091, China
-
Received Date:
2010-04-26
-
Abstract
Non-linear models with heteroscedasticity are commonly used in forestry modeling, and logarithmic regression and weighted regression are usually employed to estimate the parameters. Using the single-tree biomass data of large samples, the bias correction in logarithmic regression and comparison with weighted regression for non-linear models are studied in this paper. The immanent cause producing bias in logarithmic regression is analyzed, and a new correction factor is presented with which three commonly used bias correction factors are examined together, and the results show that the correction factors presented here and by Baskerville (1972) should be recommended which could insure the corrected model to be asymptotically consistent with that fitted by weighted regression. Secondly, the fitting results of weighted regression for non-linear models, using the weight function based on residual errors of the model estimated by ordinary least squares (OLS) and the general weight function (W=1/f(x)2) presented by Zeng (1998) respectively, are compared with each other that show two weights works well and the general function is more applicable. It is suggested that the best way to fit non-linear models with heteroscedasticity would be using weighted regression, and when the total relative error of the estimates from the model fitted by the general weight function is more than a special limit such as ±3%, a better weight function based on residual errors of the model fitted by OLS should be used in weighted regression.
-
-
References
[1]
|
Finney D J. On the distribution of a variate whose logarithm is normally distributed [J]. J R Statist Soc, Suppl. 1941,7:155-161 |
[2]
|
Baskerville G L. Use of logarithmic regression in the estimation of plant biomass [J]. Can J For Res, 1972,2:49-53 |
[3]
|
Flewelling J W, Pienaar L V. Multiplicative regression with lognormal errors [J]. For Sci, 1981, 27(2):281-289 |
[4]
|
Snowdon P. A ratio estimator for bias correction in logarithmic regressions [J]. Can J For Res, 1991, 21(5): 720-724 |
[5]
|
Parresol B R. Assessing tree and stand biomass: a review with examples and, critical comparisons [J].For Sci,1999,45(4): 573-593 |
[6]
|
Parresol B R. Additivity of nonlinear biomass equations [J]. Can J For Res, 2001,31: 865-878 |
[7]
|
曾伟生,骆期邦.论林业数表模型的研建方法[J].中南林业调查规划, 2001,20(2):1-4
|
[8]
|
王仲锋.森林生物量建模与精度分析 .北京: 北京林业大学,2006
|
[9]
|
Wiant Jr H V, Harner E J. Percent bias and standard error in logarithmic regression [J]. For Sci, 1979,25(1):167-168 |
[10]
|
Sprugel D G. Correcting for bias in log-transformed allometric equations [J]. Ecology, 1983,64(1):209-210 |
[11]
|
Lehtonen A, Mäkipää R, Heikkinen J, et al. Biomass expansion factors (BEFs) for Scots pine, Norway spruce and birch according to stand age for boreal forests [J]. Forest Ecology and Management, 2004,188:211-224 |
[12]
|
Fatemi F R. Aboveground biomass and nutrients in developing northern hardwood stands in New Hampshire, USA . USA: College of Environmental Science and Forestry, State University of New York, 2007 |
[13]
|
Beauchamp J J,Olson J S. Corrections for bias in regression estimates after logarithmic transformation [J]. Ecology, 1973,54(6):1403-1407 |
[14]
|
曾伟生.再论加权最小二乘法中权函数的选择[J].中南林业调查规划, 1998,17(3):9-11
|
[15]
|
曾伟生,骆期邦,贺东北.论加权回归与建模[J].林业科学, 1999,35(5):5-11
|
[16]
|
张会儒,唐守正,胥 辉.关于生物量模型中的异方差问题[J].林业资源管理, 1999(1):46-49
|
[17]
|
胥 辉.生物量模型方差非齐性研究[J].西北林学院学报, 1999,19(2):73-77
|
[18]
|
Zabek L M, Prescott C E. Biomass equations and carbon content of aboveground leafless biomass of hybrid poplar in Coastal British Columbia [J]. Forest Ecology and Management, 2006,223:291-302 |
-
-
Proportional views
-