Probit regression to estimate the physiological potential of hybrid maize seed

This work was carried out to study the physiological potential of artificially aged seed lots of maize. The specific aim of this study was to fit a simplified equation from Andreoli, , and present a methodology using probit regression analysis, given by the equation . We used seeds from three lots of the maize hybrid OC 705 which were submitted to the accelerated aging test, at the temperature of 43 ºC, every 24 hours. The simplified equation did not provide a good fit to the data, with r2 of at most 92%. Pearson's Chi-square test and the log-likelihood ratio Chi-square test indicated that probit regression had a good fit to the data, providing estimated values with high accuracy. It was observed that lot three maintained the highest vigor throughout the storage period.

本研究旨在探究人工老化玉米种子批的生理潜力。本研究的具体目标为拟合Andreoli提出的简化方程，并提出一套基于概率单位回归分析（probit regression analysis）的方法，该方法由对应方程给出。本研究选取玉米杂交种OC 705的3个种子批，将其置于43℃环境下开展加速老化试验（accelerated aging test），每24小时进行一次试验操作。该简化方程对试验数据的拟合效果不佳，其决定系数（r²）最高仅为92%。皮尔逊卡方检验（Pearson's Chi-square test）与对数似然比卡方检验（log-likelihood ratio Chi-square test）结果表明，概率单位回归分析对试验数据的拟合效果良好，所得估计值精度较高。观测结果显示，第3号种子批在整个贮藏期内始终保持最高的种子活力水平。