Stochastic frontiers or regression quantiles for estimating the self-thinning surface in higher dimensions?

Abstract

Stochastic frontier analysis and quantile regression are the two econometric approaches that have been commonly adopted in the determination of the self-thinning boundary line or surface in two and higher dimensions since their introduction to the field some 20 years ago. However, the rational for using one method over the other has, in most cases, not been clearly explained perhaps due to a lack of adequate appreciation of differences between the two approaches for delineating the self-thinning surface. Without an adequate understanding of such differences, the most informative analysis may become a missed opportunity, leading to an inefficient use of data, weak statistical inferences and a failure to gain greater insight into the dynamics of plant populations and forest stands that would otherwise be obtained. Using data from 170 plot measurements in even-aged Larix olgensis (A. Henry) plantations across a wide range of site qualities and with different abundances of woody weeds, i.e. naturally regenerated non-crop species, in northeast China, this study compared the two methods in determining the self-thinning surface across eight sample sizes from 30 to 170 with an even interval of 20 observations and also over a range of quantiles through repeated random sampling and estimation. Across all sample sizes and over the quantile range of 0.90 ≤  τ ≤ 0.99, the normal-half normal stochastic frontier estimation proved to be superior to quantile regression in statistical efficiency. Its parameter estimates had lower degrees of variability and correspondingly narrower confidence intervals. This greater efficiency would naturally be conducive to making statistical inferences. The estimated self-thinning surface using all 170 observations enveloped about 96.5% of the data points, a degree of envelopment equivalent to a regression quantile estimation with a τ of 0.965. The stochastic frontier estimation was also more objective because it did not involve the subjective selection of a particular value of τ for the favoured self-thinning surface from several mutually intersecting surfaces as in quantile regression. However, quantile regression could still provide a valuable complement to stochastic frontier analysis in the estimation of the self-thinning surface as it allows the examination of the impact of variables other than stand density on different quantiles of stand biomass.

Keywords

Larix olgensis / Normal-half normal distribution / Site productivity / Woody weeds / Sample size / Quantile selection

References

[1] Aigner D, Lovell CAK, Schmit P. Formulation and estimation of stochastic frontier production models. J Econom, 1977, 6: 21-37.

[2] Andrews C, Weiskittel A, D'Amato AW, Simons-Legaard E. Variation in the maximum stand density index and its linkage to climate in mixed species forests of the North American Acadian Region. For Ecol Manag, 2018, 417: 90-102.

[3] Battese GE, Corra GS. Estimation of a production frontier model: with application to pastoral zone of eastern Australia. Aust J Agric Econ, 1977, 21: 169-179.

[4] Begon M, Townsend CR, Harper JL. Ecology: from individuals to ecosystems, 2006 4 Hoboken: Blackwell Publishing.

[5] Belsley DA. Conditioning diagnostics: collinearity and weak data in regression. Wiley series in probability, 1991, New York: Wiley.

[6] Bi H. The self-thinning surface. For Sci, 2001, 47: 361-370.

[7] Bi H. Stochastic frontier analysis of a classic self-thinning experiment. Aust Ecol, 2004, 29: 408-417.

[8] Bi H, Turvey ND. Effects of Eucalyptus obliqua (L'Herit) density on young stands of even-aged Pinus radiata (D. Don). New For, 1994, 8: 25-42.

[9] Bi H, Turvey ND. A method of selecting data points for fitting the maximum biomass-density line for stands undergoing self-thinning. Aust J Ecol, 1997, 22: 356-359.

[10] Bi H, Wan G, Turvey ND. Estimating the self-thinning boundary line as a density-dependent stochastic biomass frontier. Ecology, 2000, 81: 1477-1483.

[11] Brandt SA. Modelling and visualizing uncertainties of flood boundary delineation: algorithm for slope and DEM resolution dependencies of 1D hydraulic models. Stoch Environ Res Risk Assess, 2016, 30: 1677-1690.

[12] Bravo-Oviedo A, Condés S, Río M, Pretzsch H, Ducey MJ. Maximum stand density strongly depends on species-specific wood stability, shade and drought tolerance. Forestry, 2018, 91: 459-469.

[13] Brunet-Navarro P, Sterck FJ, Vayreda J, Martinez-Vilalta J, Mohren GM. Self-thinning in four pine species: an evaluation of potential climate impacts. Ann For Sci, 2016, 73: 1025-1034.

[14] Cade BS, Guo QF. Estimating effects of constraints on plant performance with regression quantiles. Oikos, 2000, 91: 245-254.

[15] Charru M, Seynave I, Morneau F, Rivoire M, Bontemps JD. Significant differences and curvilinearity in the self-thinning relationships of 11 temperate tree species assessed from forest inventory data. Ann For Sci, 2012, 69: 195-205.

[16] Chen YT, Wang HJ. Centered-residuals-based moment tests for stochastic frontier models. Econom Rev, 2012, 31: 625-653.

[17] Comeau PG, White M, Kerr G, Hale SE. Maximum density-size relationships for Sitka spruce and coastal Douglas-fir in Britain and Canada. Forestry, 2010, 83: 461-468.

[18] Condés S, Vallet P, Bielak K, Bravo-Oviedo A, Coll L, Ducey MJ, Pach M, Pretzsch H, Sterba H, Vayreda J, Río M. Climate influences on the maximum size-density relationship in Scots pine (Pinus sylvestris L.) and European beech (Fagus sylvatica L.) stands. For Ecol Manag, 2017, 385: 295-307.

[19] Davino C, Furno M, Vistocco D. Quantile regression: theory and applications, 2013, New York: Wiley.

[20] Deng WP, Li FR. Estimation of self-thinning line for larch plantation based on extreme values. J Nanjing For Univ, 2014, 38: 11-14. (in Chinese)

[21] Dong LH, Zhang LJ, Li FR. Developing two additive biomass equations for three coniferous plantation species in northeast China. Forests, 2016, 7: 136.

[22] Ducey MJ, Knapp RA. A stand density index for complex mixed species forests in the northeastern United States. For Ecol Manag, 2010, 260: 1613-1622.

[23] Farrell MJ. The measurement of productive efficiency. J R Stat Soc A, 1957, 120: 253-281.

[24] Forsund FR, Lovell CAK, Schmidt P. A survey of frontier production functions and of their relationship to efficiency measurement. J Econom, 1980, 13: 5-25.

[25] Fried H, Lovell CAK, Schmidt S. The measurement of productive efficiency and productivity change, 2008, New York: Oxford University Press.

[26] García O. Estimating top height with variable plot sizes. Can J For Res, 1998, 28: 1509-1517.

[27] García O, Batho A. Top height estimation in lodgepole pine sample plots. West J Appl For, 2005, 20: 64-68.

[28] Ge FX, Zeng WS, Ma W, Meng JH. Does the slope of the self-thinning line remain a constant value across different site qualities? An Implication for Plantation Density Management. Forests, 2017, 8: 355.

[29] Gorham E. Shoot height, weight and standing crop in relation to density of monospecific plant stands. Nature, 1979, 279: 148.

[30] Greene W. Pesaran H, Schmidt P. Frontier production functions. Handbook of applied econometrics: microeconomics, 1997, Oxford: Blackwell 81 166

[31] Guo X, Li GR, McAleer M, Wong WK. Specification testing of production in a stochastic frontier model. Sustainability, 2018, 10: 3082.

[32] Hao L, Naiman DQ (2007) Quantile regression. Sage publications series: quantitative applications in the social sciences, vol 149. Thousand Oaks, California, 126 pp

[33] Harper J. Population biology of plants, 1977, London: Academic Press.

[34] He XM. A conversation with roger Koenker. Int Stat Rev, 2017, 85: 46-60.

[35] Herberich MM, Gayler S, Anand M, Tielbörger K. Biomass-density relationships of plant communities deviate from the self-thinning rule due to age structure and abiotic stress. Oikos, 2020

[36] Horrace WC, Parmeter CF. A Laplace stochastic frontier model. Econom Rev, 2018, 37: 260-280.

[37] Kimsey MJ Jr, Shaw TM, Coleman MD. Site sensitive maximum stand density index models for mixed conifer stands across the Inland Northwest. For Ecol Manag, 2019, 433: 396-404.

[38] Kira T, Ogasawa H, Sakazaki N. Intraspecific competition among higher plants. I. Competition-yield-density interrelationship in regularly dispersed populations. J Inst Polytech Osaka City Univ, 1953, D4: 1-16.

[39] Kocherginsky M, He XM, Mu YM. Practical confidence intervals for regression quantiles. J Comput Graph Stat, 2005, 14: 41-55.

[40] Koenker R. Quantile regression, 2005, Cambridge: Cambridge University Press.

[41] Koenker R. Quantile regression: 40 years on. Annu Rev Econ, 2017, 9: 155-176.

[42] Koenker R, Basset G. Regression quantiles. Econometrica, 1978, 46: 33-50.

[43] Koenker R, Hallock K. Quantile regression: an introduction. J Econ Perspect, 2001, 15: 43-56.

[44] Koyama H, Kira T. Intraspecific competition among higher plants VIII. Frequency distribution of individual plant weight as affected by the interaction between plants. J Inst Polytech Osaka City Univ, 1956, D7: 73-94.

[45] Kumbhakar SC, Lovell CAK. Stochastic frontier analysis, 2000, Cambridge: Cambridge University Press.

[46] Kumbhakar SC, Parmeter CF, Zelenyuk V (2018) Stochastic frontier analysis: foundations and advances. Working paper series no. WP02/2018, School of Economics, University of Queensland, Australia

[47] Kweon D, Comeau PG. Effects of climate on maximum size-density relationships in Western Canadian trembling aspen stands. For Ecol Manag, 2017, 406: 281-289.

[48] Long JN, Shaw JD. A density management diagram for even-aged Sierra Nevada mixed-conifer stands. West J Appl For, 2012, 27: 187-195.

[49] Marchi M. Nonlinear versus linearised model on stand density model fitting and stand density index calculation: analysis of coefficients estimation via simulation. J For Res, 2019, 30: 1595-1602.

[50] Meesters A. A note on the assumed distributions in stochastic frontier models. J Prod Anal, 2014, 42: 171-173.

[51] Meeusen W, van den Broeck J. Efficiency estimation from Cobb–Douglas production functions with composed error. Int Econ Rev, 1977, 8: 435-444.

[52] Montigny LD, Nigh G. Density frontiers for even-aged Douglas-fir and western hemlock stands in coastal British Columbia. For Sci, 2007, 53: 675-682.

[53] Norberg RA. Theory of growth geometry of plants and self-thinning of plant populations: geometric similarity, elastic similarity, and different growth modes of plant parts. Am Nat, 1988, 131: 220-256.

[54] Parmeter CF, Kumbhakar SC. Efficiency analysis: a primer on recent advances. Found Trends® Econom, 2014, 7: 191-385.

[55] Peng W, Pukkala T, Jin XJ, Li FR. Optimal management of larch (Larix olgensis A. Henry) plantations in Northeast China when timber production and carbon stock are considered. Ann For Sci, 2018, 75: 63.

[56] Pretzsch H. A unified law of spatial allometry for woody and herbaceous plants. Plant Biol, 2002, 4: 159-166.

[57] Pretzsch H. Forest dynamics, growth, and yield, 2009, Heidelberg: Springer, Berlin 664

[58] Pretzsch H, Matthew C, Dieler J. Matyssek R, Schnyder H, Oßwald W, Ernst D, Munch JC, Pretzsch H. Allometry of tree crown structure. Relevance for space occupation at the individual plant level and for self-thinning at the stand level. Growth and defence in plants, vol 220, 2012, Berlin: Springer 287 310 (Ecological Studies)

[59] Puettmann KJ, Hibbs DE, Hann DW. The dynamics of mixed stands of Alnus rubra and Pseudotsuga menziesii: extension of size-density analysis to species mixture. J Ecol, 1992, 80: 449-458.

[60] Quiñonez-Barraza G, Tamarit-Urias JC, Martínez-Salvador M, GarcíaCuevas X, de los Santos-Posadas HM, Santiago-García W. Maximum density and density management diagram for mixed-species forests in Durango, Mexico. Revista Chapingo Serie Ciencias Forestales y del Ambiente, 2018, 24: 73-90.

[61] Reineke LH. Perfecting a stand-density index for even-aged forests. J Agric Res, 1933, 46: 627-638.

[62] Reyes-Hernandez V, Comeau PG, Bokalo M. Static and dynamic maximum size-density relationships for mixed trembling aspen and white spruce stands in western Canada. For Ecol Manag, 2013, 289: 300-311.

[63] Riofrío J, Del Río M, Bravo F. Mixing effects on growth efficiency in mixed pine forests. Forestry, 2016, 90: 381-392.

[64] Ritter C, Simar L. Pitfalls of normal-gamma stochastic frontier models. J Prod Anal, 1997, 8: 167-182.

[65] Rivoire M, Le Moguedec G. A generalized self-thinning relationship for multi-species and mixed-size forests. Ann For Sci, 2012, 69: 207-219.

[66] Rose R, Rosner LS, Ketchum JS. Twelfth-year response of Douglas-fir to area of weed control and herbaceous versus woody weed control treatments. Can J For Res, 2006, 36: 2464-2473.

[67] Rosenberg AS, Knuppe AJ, Braumoeller BF. Unifying the study of asymmetric hypotheses. Polit Anal, 2017, 25: 381-401.

[68] Salas-Eljatib C, Weiskittel AR. Evaluation of modeling strategies for assessing self-thinning behavior and carrying capacity. Ecol Evol, 2018, 8: 10768-10779.

[69] SAS Institute Inc.. SAS/STAT^® 13.1 user’s guide, 2013, Cary: SAS Institute Inc.

[70] SAS Institute Inc.. SAS/ETS^® 13.2 user’s guide, 2014, Cary: SAS Institute Inc.

[71] Schmidt P. Frontier production functions. Econom Rev, 1985, 4: 289-328.

[72] Shinozaki H, Kira T. Intraspecific competition among higher plants VII. Logistic theory of the C-D effect. J Inst Polytech Osaka City Univ, 1956, D7: 35-72.

[73] Socha J, Zasada M. Stand density and self-thinning dynamics in young birch stands on post-agricultural lands. Sylwan, 2014, 158: 340-351.

[74] Solomon DS, Zhang LJ. Maximum size–density relationships for mixed softwoods in the northeastern USA. For Ecol Manag, 2002, 155: 163-170.

[75] Stevenson R. Likelihood functions for generalized stochastic frontier estimation. J Econom, 1980, 13: 58-66.

[76] Sun HG, Zhang GJ, Duan AG. A comparison of selecting data points and fitting coefficients methods for estimating self-thinning boundary line. Chin J Plant Ecol, 2010, 34: 409-417. (in Chinese)

[77] Tarr G. Small sample performance of quantile regression confidence intervals. J Stat Comput Simul, 2012, 82: 81-94.

[78] Vospernik S, Sterba H. Do competition-density rule and self-thinning rule agree?. Ann For Sci, 2015, 72: 379-390.

[79] Wagner F, Rutishauser E, Blanc L, Herault B. Effects of plot size and census interval on descriptors of forest structure and dynamics. Biotropica, 2010, 42: 664-671.

[80] Wang WS, Amsler C, Schmidt P. Goodness of fit tests in stochastic frontier models. J Prod Anal, 2011, 35: 95-118.

[81] Weiskittel A, Kuehne C. Evaluating and modelling variation in site-level maximum carrying capacity of mixed-species forest stands in the Acadian Region of northeastern. N Am For Chron, 2019, 95: 171-182.

[82] Weiskittel A, Gould P, Temesgen H. Sources of variation in the self-thinning boundary line for three species with varying levels of shade tolerance. For Sci, 2009, 55: 84-93.

[83] Westoby M. The self-thinning rule. Adv Ecol Res, 1984, 14: 167-225.

[84] White J. Solbrig OT. Demographic factors in populations of plants. Demography and evolution in plant populations, 1980, Berkeley: University of California Press 21 48

[85] White J. The allometric interpretation of the self-thinning rule. J Theor Biol, 1981, 89: 475-500.

[86] White J. White J. The thinning rule and its application to mixtures of plant populations. Studies on plant demography, 1985, London: Academic Press 291 309

[87] White J, Harper JL. Correlated changes in plant size and number in plant populations. J Ecol, 1970, 58: 467-485.

[88] Whittington R. Laying down the –3/2 power law. Nature, 1984, 311: 217.

[89] Woodall CW, Fiedler CE, Milner KS. Stand density index in uneven-aged ponderosa pine stands. Can J For Res, 2003, 33: 96-100.

[90] Woodall CW, Miles PD, Vissage JS. Determining maximum stand density index in mixed-species stands for strategic-scale stocking assessments. For Ecol Manag, 2005, 216: 367-377.

[91] Xue L, Hou XL, Li QJ, Hao YT. Self-thinning lines and allometric relation in Chinese fir (Cunninghamia lanceolata) stands. J For Res, 2015, 26(2): 281-290.

[92] Yoda K, Kira T, Hozumi K. Intraspecific competition among higher plants IX. Further analysis of the competitive interaction between adjacent individuals. J Inst Polytech Osaka City Univ, 1957, D8: 161-178.

[93] Yoda K, Kira T, Ogawa H, Hozumi K. Self-thinning in overcrowded pure stands under cultivated and natural conditions (Intraspecific competition among higher plants XI). J Biol Osaka City Univ, 1963, 14: 107-129.

[94] Yu B. Artifical cultivation techniques of Larix olgensis. Heilongjiang Sci Technol Inf, 2008, 56: 55-57. (in Chinese)

[95] Zhang LJ, Bi HQ, Gove JH, Heath LS. A comparison of alternative methods for estimating the self-thinning boundary line. Can J For Res, 2005, 35: 1507-1514.

[96] Zhang JW, Oliver WW, Powers RF. Reevaluating the self-thinning boundary line for ponderosa pine (Pinus ponderosa) forests. Can J For Res, 2013, 43: 963-971.

[97] Zheng LS, Li HK. Effects of model form and region on prediction for aboveground biomass of Larix. For Resour Manag, 2013, 2: 83-88. (in Chinese)

[98] Zhou ML, Lei XD, Duan GS, Lu J, Zhang HR. The effect of the calculation method, plot size, and stand density on the top height estimation in natural spruce-fir-broadleaf mixed forests. For Ecol Manag, 2019, 453: 117574.