Econometric techniques

In the case of chronic processes, the patient can, at any time, take the decision to stop the treatment, as a reaction to changes in price or other variables. Whereas the econometric technique for analysing the demand for pharmaceuticals for acute processes is that of discrete choice models, here we will apply duration or survival models. In addition, the technology of drags for chronic diseases can have non-constant returns to scale. For example, the consumption of anxiolytics raises the tolerance and reduces the effect, resulting in an increase in the necessary dose. 

Natural gas is sold in regional markets defined by pipeline capacity. Because of the importance of Texas and California in today s hydrogen economy, Figure 4 presents the data for the West Texas Intermediate Crude Oil Spot Price and City Gate Natural Gas Prices in Texas and in California. By applying time-series econometric techniques to these data, one can conclude that the price of oil leads natural gas prices and that the price of oil does not follow changes in any other energy price [9,10]. 

In general, policy analysts use two structural types of models to make forecasts of future activity levels econometric and engineering or process models. Econometric models use historic data and relationships to estimate future trends in variables of interest. Engineering or process models use the physical relationships of production processes (i.e., the relationship between inputs to a production process and its outputs) to predict levels of the dependent variables. In general, the use of econometric modeling techniques to forecast activity levels provides a better long-term (beyond 20 years) trend because it relies on long-term, historical... 

Partial Least Squares Regression (PLS) is a multivariate calibration technique, based on the principles of Latent Variable Regression. Originated in a slightly different form in the field of econometrics, PLS has entered the spectroscopic scene.46,47,48 It is mostly employed for quantitative analysis of mixtures with overlapping bands (e.g. mixture of glucose, fructose and sucrose).49,50... 

Inflation modelling for probabilistic analysis has to deal with the properties of inflation datasets. These datasets follow an exponential trendline, with a heteroskedastic deviation. Furthermore, the dataset has positive serial autocorrelation and some indexes, like gasoil and bitumen, are strongly positively correlated. These properties are included with a modelling technique used in the field of econometrics, named ARIMA-modelling. 

---