Foster Ken, Anwar Nasim, Pogue Rhea, Morré Dorothy M, Keenan T W, Morré D James
Department of Agricultural Economics, Department of Medicinal Chemistry and Molecular Pharmacology.
Nonlinearity Biol Toxicol Med. 2003 Jan;1(1):51-70. doi: 10.1080/15401420390844465.
Seasonal decomposition analyses were applied to the statistical evaluation of an oscillating activity for a plasma membrane NADH oxidase activity with a temperature compensated period of 24 min. The decomposition fits were used to validate the cyclic oscillatory pattern. Three measured values, average percentage error (MAPE), a measure of the periodic oscillation, mean average deviation (MAD), a measure of the absolute average deviations from the fitted values, and mean standard deviation (MSD), the measure of standard deviation from the fitted values plus R-squared and the Henriksson-Merton p value were used to evaluate accuracy.Decomposition was carried out by fitting a trend line to the data, then detrending the data if necessary, by subtracting the trend component. The data, with or without detrending, were then smoothed by subtracting a centered moving average of length equal to the period length determined by Fourier analysis. Finally, the time series were decomposed into cyclic and error components. The findings not only validate the periodic nature of the major oscillations but suggest, as well, that the minor intervening fluctuations also recur within each period with a reproducible pattern of recurrence.
季节分解分析被应用于对具有24分钟温度补偿周期的质膜NADH氧化酶活性的振荡活性进行统计评估。分解拟合用于验证循环振荡模式。使用三个测量值,即平均百分比误差(MAPE),一种周期性振荡的度量;平均绝对偏差(MAD),一种与拟合值的绝对平均偏差的度量;以及平均标准偏差(MSD),与拟合值的标准偏差的度量,再加上R平方和亨里克森 - 默顿p值来评估准确性。分解是通过对数据拟合一条趋势线,然后如有必要通过减去趋势成分对数据进行去趋势化来进行的。有或没有去趋势化的数据,然后通过减去长度等于由傅里叶分析确定的周期长度的中心移动平均值来进行平滑处理。最后,时间序列被分解为循环和误差成分。这些发现不仅验证了主要振荡的周期性,还表明,次要的中间波动在每个周期内也以可重复的复发模式再次出现。