Suchergebnisse

Cover -- Title page -- Copyright and Credits -- Contributors -- Table of Contents -- Preface -- Section 1: Getting Started -- Chapter 1: The History and Development of Time Series Forecasting -- Understanding time series forecasting -- The problem with dependent data -- Moving average and exponential smoothing -- ARIMA -- ARCH/GARCH -- Neural networks -- Prophet -- Summary -- Chapter 2: Getting Started with Facebook Prophet -- Technical requirements -- Installing Prophet -- Installation on macOS -- Installation on Windows -- Installation on Linux -- Building a simple model in Prophet -- Interpreting the forecast DataFrame -- Understanding components plots -- Summary -- Section 2: Seasonality, Tuning, and Advanced Features -- Chapter 3: Non-Daily Data -- Technical requirements -- Using monthly data -- Using sub-daily data -- Using data with regular gaps -- Summary -- Chapter 4: Seasonality -- Technical requirements -- Understanding additive versus multiplicative seasonality -- Controlling seasonality with Fourier order -- Adding custom seasonalities -- Adding conditional seasonalities -- Regularizing seasonality -- Global seasonality regularization -- Local seasonality regularization -- Summary -- Chapter 5: Holidays -- Technical requirements -- Adding default country holidays -- Adding default state/province holidays -- Creating custom holidays -- Creating multi-day holidays -- Regularizing holidays -- Global holiday regularization -- Individual holiday regularization -- Summary -- Chapter 6: Growth Modes -- Technical requirements -- Applying linear growth -- Understanding the logistic function -- Saturating forecasts -- Increasing logistic growth -- Non-constant cap -- Decreasing logistic growth -- Applying flat growth -- Summary -- Chapter 7: Trend Changepoints -- Technical requirements -- Automatic trend changepoint detection.

Introduction: Forecasting in peace research / Håvard Hegre, Nils W Metternich

Presents a method for determing outcomes of political/economic events, based on power structure of the conflict. Analyzes economic and political issues resolved in national or international setting using the Probabilistic Conflict Forecasting Method.

Summary This work explores the possibilities of forecasting German gas imports and German gas consumption in detail by: • Identifying proper terminology, with initial thoughts on the status quo of forecasting, using analogies with music (section 3.1), • Understanding forecasting models based on statistics and machine learning (section 3.2), • Applying domain knowledge while constructing data sets (data acquisition) - various models have been tested on low resolution data sets (monthly and yearly data). In classical forecasting, forecasts are conducted at company level (e.g. electricity load forecasting) for a short-term period fulfilling the requirement of profit maximization. In contrast, this work's forecasts spatially cover 1) all import nodes into Germany, 2) gas consumption in Germany with the focus on energy security. • Examining various notions of complexity in relation to forecasting Novelty in methodology • Considering infrastructure as the upper limit for the forecasting model (section 4.1) • Testing models for ex post forecasting of gas imports into Germany (section 4.2) • Testing models for ex post forecasting of gas consumption in Germany (section 4.3). Producing an ex ante forecast by answering "what will be the future gas consumption in Germany in next ten years" would cause the forecasting error to increase dramatically due to the uncertainty of future input values, such as population (lower risk) or, weather (higher uncertainty), which would lead to prediction intervals being too wide to make any statement about the future imports or consumption. • Applying complexity measures such as approximate entropy ApEn and sample entropy to two self-constructed real-world data sets on gas imports to Germany and gas consumption in Germany (section 3.3) ; Diese Arbeit untersucht die Möglichkeiten zur Prognose von Deutschen Gasimporten sowie des Gasverbrauchs in Deutschland im Detail, wie folgt: • Identifikation klarer Terminologie, mit ersten Gedanken zum Status Quo der Prognosetechnik (Abschnitt 3.1) • ...

To forecast an election means to declare the outcome before it happens. Scientific approaches to election forecasting include polls, political stock markets & statistical models. I review these approaches, with an emphasis on the last, since it offers more lead time. Consideration is given to the history & politics of statistical forecasting models of elections. Rules for evaluating such models are offered. Examples of actual models come from the United States, France & the United Kingdom, where this work is rather new. Compared to other approaches, statistical modeling seems a promising method for forecasting elections. 1 Table, 51 References. Adapted from the source document.