stepwise regression analysis

A proper planting date in canola allows the crop to grow sufficiently and minimizes the damaging effects of stress. On the other hand, early autumn rains and the problem of timely farm preparation and unavailability of farm fields due to summer crops such as peanuts in the Moghan region cause delays in planting dates for autumn crops such as canola. Problems such as supplying primary soil water for crop establishment of canola fields, pest damage such as flea beetles in delayed cultivation, the possibility of damage from cold and frost stress, as well as the decline in yield due to canola exposure to drought and heat stress due to delayed planting are the main problems of canola cultivation in the Moghan region. Because of this, canola transplanting can solve the problems and be a good choice in these kinds of situations.

In order to agronomic evaluation of direct cultivation and transplantation of spring canola in different plant densities under delayed conditions in the Moghan region, an experiment was conducted as a randomized complete block design with three replications in two cropping years, 2019-2020 and 2020-2021. Experimental treatments included direct sowing of seeds as a control at a rate of 6 kg/ha; transplanting with densities of 20, 30, and 40 plants per square meter, each density with both bare-root transplant and potted-root transplant; and also with one seedling or two seedlings in the planting hole. During the experiment, traits including flowering initiation, flowering completion, flowering period, growth period, plant height, first pod height, stem diameter, branch number, pod length, pod thickness, pod number, seed number per pod, 1000-seed weight, and seed yield were recorded. Correlations among traits and path analysis were performed to investigate seed yield's direct and indirect effects.

The results showed that transplanting on average caused a significant decrease in phenological traits such as flowering initiation (145.5 GDD), flowering completion (207.8 GDD), growth period (158.9 GDD), and increased flowering period (156.6 GDD) during the experimental years. Also, transplanting caused a significant increase in stem diameter (2.8 mm), branch number (2.3), pod diameter (0.6 mm), number of pods per plant (158.2), number of seeds per pod (3.3), 1000-seed weight (0.69 g) and seed yield (1894.4 kg/ha). In control treatment as delayed direct sowing, seed yield in total experimental years showed a yield decline of 62.2% (1150.7 kg/ha) compared to transplanting cultivation treatments (3045.1 kg/ha). Comparing the bare-root transplant treatments, it was found that 40 plants per square meter had a higher number of pods per plant, 1000-seed weight, and seed yield than the 30 and 20 treatments and had a significant difference with them. Phenotypic correlation among agronomic traits showed seed yield had a negative and significant correlation with phenological traits: flowering initiation (-0.90**), flowering completion (-0.90**), and growth period (-0.77**) with a positive and non-significant effect on the flowering period (0.52). The results of stepwise regression analysis showed that the number of pods per plant (0.241*) and pod thickness (0.229*) had the most significant direct and positive effects on seed yield, respectively, while flowering completion (-0.559**) had the most negative and significant direct effect on seed yield.

Under delayed conditions, bare-root transplanting of canola with 40 plants per square meter density and one seedling in the planting hole was superior in terms of seed yield and phenological traits compared to potted-root transplanting and direct cultivation and is recommended in Moghan region.

Path and biplot analysis based on linear and bilinear models, respectively, are of the effective statistical methods for expressing relationships between traits in plant. In order to study of effective traits on seed yield of bitter vetch, 20 ecotypes were evaluated in a randomized block design with three replications during 2017-2018 growing season in the Research Field of Faculty of Agriculture at Vali-e-Asr University of Rafsanjan. The analysis of variance revealed significant differences among the ecotypes for 16 studied traits except plant and root height in maturity stage and one seed pod length suggesting high variation among the ecotypes for the majority of the studied traits. Correlation analysis showed that there were a positive and significant correlation between seed yield and the most of traits except phenological traits including days to 50% flowering and days to maturing start and days to 50% maturing and two morphological traits including root height in maturity stage and two seed pod length. Stepwise regression analysis indicated that thousand kernel weight, number of seed per plant and number of lateral branch entered into model and explained about 65.7 of the total variation. The results obtained from path analysis demonstrated that thousand kernel weight had the most direct effects on seed yield (an increase of 0.657 g in seed yield per 1 g increase in thousand kernel weight). Biplot of first two components divided the successful ecotypes into two groups and identified the most effective traits association with the success of each group.

Inequality of increasing population and food will be worse as soon as possible, because the human population is expected to increase to over 8 billion by the year 2020 (Ashraf and Harris, 2006). Therefore, plant breeding is a permanent and economic method to solving the problem. Detection of effective traits on grain yield is an efficient approach to increase crop production. Among agronomy crops, chickpea plays vary important role in human life. KHalighi et al. (2011) showed that grain yield had a significant correlation with pod number per plant, seed number per plant, harvest index and biomass. In another research (Fayyaz and Talebi, 2009), the traits of number of pod and number of seed per plant were proposed as the most important traits for yield increasing in chickpea. Yucel et al. (2006) suggested that seed number per plant and pod number per plant have maximum direct effect on grain yield. Stepwise regression showed that the traits of biomass, harvest index and 100-kernel weight had great effect on chickpea grain yield, so the traits explained the main part of yield variance (Mardi et al, 2003). Detection and identifying the most effective traits on chickpea grain yield was the main objective of the current research.
The experiment was conducted based on randomized complete block design with three replications on 18 chickpea promising lines and two cultivars (as control) under normal moisture condition in 2011-2012. The experiment was conducted in agricultural and natural resources research center of Kurdestan, Iran. The traits of day to 50 percent flowering, day to 50 percent podding, day to maturity, relative water content, number of sub-branches, number of main branches, seed number per plant, pod number per plant, plant height, 100- kernel weight, grain yield, harvest index and biomass were measured. Correlation analysis, stepwise regression analysis, and path analysis were used for identifying the best genotypes and traits in the current experiment. Phenotypic coefficient variation, genotypic coefficient variation and broad sense heritability were estimated for all traits.
The results showed that the traits of number of sub-branch per plant, number of pod per plant, number of main-branch per plant, seed number per plant, harvest index and biomass had the maximum, positive and significant correlation with grain yield, while the trait of day to 50% flowering had the maximum, negative and significant correlation with grain yield. Stepwise regression results indicated that the traits of number of sub-branch, biomass, harvest index and day to 50% flowering (with negative beta) were the most important effective traits on the mean of grain yield and explained 97.98% of yield variance. Path analysis results confirmed the obtained results. The path analysis results showed that the traits of number of sun-branch and biomass had the maximum positive direct effect on grain yield. 100-kerenl weight and plant height had the highest and lowest heritability respectively. The maximum genotypic coefficient of variance belongs to 100-kernel weight, number of sub and main branches respectively. Meanwhile, the trait of harvest index and biomass (by increasing the number of sub-branch) had the maximum indirect positive effect on yield. As a result, increasing the sub-branch number and decreasing of day to 50% flowering may be proposed as a useful strategy for yield increase under the current experiment condition.
Grain yield is a quantitative trait with low broad sense heritability. Therefore, identifying the traits correlated with grain yield is a suitable strategy for improvement and indirect selection of grain yield. Stepwise regression and path analysis are an efficient and useful statistical approach for reaching were the above aim. According to results, the traits of number of sun-branch, number of main-branch and biomass recognized as the best suitable traits for indirect grain yield selection under normal moisture stress condition.

In pistachio rootstocks breeding, correlation of characteristics and their relationships with plan vigor is very important. Banebaghi is a natural cross of Pistacia subspecies that have a high variation in generated seedlings. In order to study these variations and relationships among morphological characteristics, an experiment was designed with 108 seedlings of Banebaghi. Fourteen growth characteristics such as seedling primary growth type, length and diameter of shoot, leaf and leaflet, dimensions of leaf and leaflet, leaf and leaflet number and two physiological parameters, including chlorophyll index and chlorophyll florescence were evaluated. Based on results, a high morphological variation was observed in Banebaghi seedlings, so that seedlings divided in two groups based on types of growth: pistachio growth type and mastic tree growth type. In both growth types, there were positively and negatively correlations among vigor of seedling and stem length, leaf number, stem diameters and chlorophyll index. Based on stepwise regression analysis, in both growth types stem volume and leaf number influenced the stem length and diameters of seedlings.

To identify the most effective traits on corn yield by using biofertilizer under water deficit stress, an experiment was carried out in split plot based on randomized complete block design with three replications in Hashtgerd region in 2012. In this experiment, water deficit stress, as the main factor, consisted of four levels (control, stress at flowering, stress at grain filling and stress at both flowering and grain filling stages). Biological fertilizers as sub-plot consisted of four levels of biological fertilizer applications (control, application of Nytrazhyn, application of Barvar 2 fertile and application of both fertilizer Nytrazhyn and Barvar 2). Traits evaluated were seed yield, harvest index, 100 seed weight, number of ears per square meter, number of rows per ear, number of seed per row, weight, diameter and length of seeds. To determine the most effective traits simple correlation, stepwise regression analysis, path analysis and principal component analysis were performed. Multivariate analysis showed that harvest index, 100 grain weight and number of ears per square meter were highly effective on grain yield. They explained 77% and 66% of the total variance under both normal and drought stress conditions respectivly. The highest direct positive effect was related to harvest index under both conditions. To improving seed yielding potential of corn under water deficite cinditions emphasis should be placed mainly on the selection of genotypes with higher harvest indices.

The Chamomile of Anthemis haussknechtii species is one of the medicinal plants which is related to the compositae family which grows as wild type in some parts of the country. Essential oils of chamomile flowers used extensively in pharmaceuticals, food, cosmetics and health. Ecological conditions have a major role in the growth, yield and quality of medicinal plants. The purpose of this study was evaluation of morphological and phonological, and qualitative traits such as: length and width of vegetative crown, plant height, flower number per plant, fresh and dry weight, time of flowering, time of harvest, essential oil percentage, essential oil yield, growth degree days during flowering and harvesting in ecological condition of Karaj twon. The experiment was conducted in a Randomized Completely Blocks Design (RCBD) with three replicates in Alborz research station. Analysis variance showed that there was significant difference between population for all traits for Anthemis haussknechtii species (p ≤0.01) except for essential oil yield which was significance at p ≤0.05. This study showed that Kohdasht and Kermanshah populations of Anthemis haussknechtii species, considered as superior population and also they flowered sooner than other populations. Result of simple correlation analysis showed that that there was significant positive correlation between essential oil yield with fresh and dry weight of foliage biomass, flower number per bush and length and width of vegetative crown and this result indicated that amount of top heading branch yield would be increased by increasing of above mentioned traits. Result of Stepwise Regression Analysis for essential oil yield as variable trait showed that yield of foliage biomass and percent of essential oil have more effective on essential oil yield.

This research which aims to survey the relation between tobacco leaf yield and its components is done by using 118 varieties of Tobacco in a research farm of Guilan Tobacco Research Institute. Regarding to five random selections of plants، Twenty three traits of each variety were evaluated. Correlation between dry matter of leaf yield with plant height، stem diameter، leaf length، leaf width، leaf area index، upper، lower & middle internodes length، days to flowering، green matter of leaf yield، green plant weight، biomass، percentage of dry matter and harvest index was positive and significant and with leaf shape index، flowering period، length of inflorescence and percentage of root note nematode infestation was negative and significant. In Stepwise regression analysis the most effective factors on dry matter of leaf yield were particulars of green matter of leaf yield، percentage of dry matter، plant height and leaf area index. According to the results of path analysis the most direct and positive effect on dry matter of leaf yield was related to green matter of leaf yield. In path analysis between green matters of leaf yield with other studied characteristics، the particulars of the green plant weight and leaf length had the most positive direct and indirect effect on green matter of leaf yield. In factor analysis، 7 independent factors explain the variation of 80. 54 percent of all data.

This study was carried out to investigate flower yield and yield components in Iranian Rosa damascena using 20 genotypes, collected from different regions of Iran and grown in Drylands and Deserts Research Station of Kashan, Iran, using a randomized complete block design with three replications in 2007-2008. Flower yield per plant, number of flowers per plant, plant height, leaf length, width, length and width of bud, length and width of pedicle, fresh weight of flower, number of petal per flower, length and width of receptacle, plant growth date, budding date, time of flowering and flowering period were recorded. A significant phenotypic correlation indicated a strong positive relationship between flower yield and number of flowers per plant. Result of stepwise regression analysis showed that 94 percent of total variation of flowers yield per plant could be explained by number of flowers per plant and weight of the fresh flower. The path analysis indicated that the number of flower per plant had the most direct effect on the yield. This suggested that number of flowers per plant and weight of the fresh flowers are the most important components determining flower yield per plant in Damask rose.