Search Results

You are looking at 1 - 10 of 110 items for :

  • stepwise regression x
  • Refine by Access: All x
Clear All
Free access

Reza Amiri, Kourosh Vahdati, Somayeh Mohsenipoor, Mohammad Reza Mozaffari, and Charles Leslie

Torrie, 1980 ). Two main traits, kernel percentage and blight susceptibility, were analyzed as dependent variables and traits influencing these were detected by stepwise regression ( Draper and Smith, 1998 ). Direct (standardized partial regression

Open access

Yun Kong, Xiangyue Kong, and Youbin Zheng

shoots. After that, five linear models with 6, 4, 3, 2, and 1 predictor variables were developed using stepwise regression. In addition, using SML or SMD as a single predictor variable, two nonlinear models (i.e., power function) and two linear models

Free access

Arthur Villordon, Christopher Clark, Tara Smith, Don Ferrin, and Don LaBonte

data set (n = 60) was used for least squares-based linear regression analysis (forward and stepwise selection; SPSS Version 15). The predictor variables included wind (speed and direction), air temperature (growing degree-days), soil temperature

Full access

Yuhung Lin and Yaling Qian

method ( Table 3 ). The Pearson correlation test was performed by Proc CORR to obtain Pearson statistic coefficients of individual minerals and its relationship with turf quality. Stepwise regression was conducted to determine whether any of the minerals

Full access

Yung-Kun Chuang, I-Chang Yang, Chao-Yin Tsai, Jiunn-Yan Hou, Yung-Huei Chang, and Suming Chen

the input for spectral analysis. Two standard multivariate analysis methods, modified partial least-squares regression (MPLSR) and stepwise multiple linear regression (SMLR), were used to explore the relationships between the reflectance spectra and

Open access

Han Xu, Cuihua Bai, Wei Wang, Changmin Zhou, Luwei Zhu, and Lixian Yao

proline (Pro)] and bitter-taste AA [the sum of Val, Met, Ile, Leu, Phe, Lys, arginine (Arg), and Pro] were classified as the method reported by Kato et al. (1989) . Multiple stepwise regression analysis was performed between foliar nutrient and pulp FAA

Free access

L. G. Albrigo

In Florida, pounds soluble solids per box (% soluble solids × % juice × weight) can be 60% higher in some years compared to the lowest years. Pounds solids, soluble solids and juice content data were obtained for the different citrus growing districts in Florida for a 20-year period from the USDA and Florida Agricultural Statistics Service. Weather data for each district was obtained from US National Weather Service records. Total rainfall and average daily temperature were calculated for 2–month periods from prior to the normal bloom period until harvest. Juice data was regressed against weather data and the previous years pounds solids using a stepwise multiple regression program. R2 values for early oranges, `Marsh Seedless' grapefruit and `Valencia' were 0.48, 0.48 and 0.72, respectively. Prebloom and bloom rainfall and temperatures were frequently positively correlated, while summer rainfall often was a negatively correlated independent variable to final pounds solids. Additional data and physiological implications will be discussed.

Free access

Arthur Villordon, Craig Roussel, and Tad Hardy

The sweetpotato weevil [SPW, Cylas formicarius (Fabricius)] is an important economic pest in “pink-tagged” or SPW-infested areas of Louisiana. From time to time, sweetpotato weevils are detected in “green-tagged” or SPW-free locations. When sweetpotato weevils are detected in “green'tagged” areas, the produce is quarantined and may not be shipped to locations that do not allow “pink-tagged” sweetpotatoes. As part of the statewide SPW monitoring program, the Louisiana Department of Agriculture and Forestry (LDAF) conducts a statewide pheromone-based trapping program to monitor SPW presence in beds and fields. We used SPW presence-absence data with a GIS-based logistic regression modeling tool to assess the feasibility of developing a model for predicting SPW risk in sweetpotato beds. Using pheromone trap data from 2001–03, we performed stepwise logistic regression experiments to assess the role of various weather variables (daily mean maximum and minimum temperature, rainfall) in the occurrence of SPW in beds. Our modeling experiments showed a strong relationship of mean daily minimum temperature during the winter months with SPW occurrence in beds. In particular, a logistic regression equation developed from 2003 trap data and mean April daily minimum temperature created a spatially accurate map of SPW risk for 2002. However, the same model did not accurately predict the 2001 SPW risk. These results indicate that additional variables are needed to improve the predictive ability of the model. Spatial risk mapping can be a potentially useful tool for decision makers in choosing between risk-averse and -prone decisions.

Free access

S.B. Sterrett, M.R. Henningre, and G.S. Lee

acknowledge the assistance of C.P. Savage, Jr. and F.W. Punk with the field studies and M. Lentner with the regression analyses. The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper

Free access

Louise Ferguson, Hesham Gawad, G. Steven Sibbett, Mark Freeman, and James J. Hatakeda

A stepwise multiple regression analysis, using payment by processors as the dependent variable (Y) and numerous physical and chemical characteristics as the independent variables (X), demonstrated that the primary factor determining `Manzanillo' olive (Olea europaea L.) value at harvest was size. Optimal crop value correlated strongly with the combined percentage of standard, medium, large, and extra-large olives; R' values were 0.93***, 0.93***, and 0.42 (ns) in 1984, 1985, and 1986, respectively. As the harvest season progressed, increased percentages of olives within these size classifications, not weight increases of individual olives within the size categories, produced the increase in value. Individual olives within size categories maintained the same weight through the harvest season, regardless of tree crop load. The best criterion for predicting optimal harvest time “is the total percentage of standard, medium, large, and extra-large olives.