Time Series Analysis: Forecasting and Control
Time Series Analysis: Forecasting and Control
Time Series Analysis, Forecasting and Control
Time Series Analysis, Forecasting and Control
Hi-index | 0.00 |
This article analyses scientific growth time series using data for Spanish doctoral theses from 1848 to 2009, retrieved from national databases and an in-depth archive search. Data are classified into subseries by historical periods. The analytical techniques employed range from visual analysis of deterministic graphs to curve-fitting with exponential smoothing and AutoRegressive Integrated Moving Average models. Forecasts are made using the best model. The main finding is that Spanish output of doctoral theses appears to fit a quasi-logistic growth model in line with Price's predictions. An additional control variable termed year-on-year General Welfare is shown to modulate scientific growth, especially in the historical period from 1899 to 1939.