The R Markdown report will focus on analyzing, modeling, and forecasting time series data for the Morocco Stock Index 20 (MSI 20). It aims to compare classical methods, particularly ARIMA with the Box-Jenkins methodology, with advanced Machine Learning techniques like Support Vector Regression (SVR) and Multi-Layer Perceptron (MLP). The report will delve into the details of the Box-Jenkins methodology, emphasizing key steps such as identification, estimation, and model validation. The analysis will highlight the importance of differencing, order selection (p, d, q), and residual verification for ARIMA model validity. This comprehensive approach aims to provide a detailed overview of various time series modeling methodologies, elucidating their advantages and limitations. Additionally, the report will establish connections between traditional methods and advanced techniques in the analysis and forecasting of the MSI 20 closing price time series.

This tutorial aims to explore the monthly evolution of total crude oil and petroleum products in thousands of barrels in the United States from January 01, 2000, to September 01, 2023, over 23 years. After a brief introduction to our series, we will apply distinct approaches to each of them. The Box Jenkins methodology will be used to analyze the series. The main objective of this study is to generate forecasts for the next 58 months, using the complete data set to build the respective forecast models.

This project explores the application of Generalized Linear Models (GLM) for calculating pure premium in non-life insurance. By modeling the relationships between independent variables (rating factors) and a dependent variable (claim frequency or claim cost), GLMs offer a statistical framework for premium estimation. The objectives encompass data cleansing, variable selection, distribution and link function choices, model construction, parameter estimation, model evaluation, and the pricing of new insurance policies. The study concludes with a final report summarizing the entire process.

This project focuses on the comprehensive analysis and modeling of time series data, employing a robust toolkit that incorporates regression models, sophisticated graphs, and forecasting techniques. Through the utilization of advanced statistical methods, the study aims to delve into intricate patterns within time-dependent datasets, providing valuable insights for forecasting future trends. The forecaster's toolbox, enriched with regression models, serves as a versatile instrument for uncovering relationships, visualizing data dynamics, and making informed predictions in the realm of time series analysis.

In this study, we employ statistical inference techniques to develop econometric models based on economic data. These models, rooted in the assumption that economic agents optimize their behavior, are used to compare the performance of various assets, including 'green global,' 'green US gov,' and 'green US corporate,' against the broader stock market represented by 'xle' and 'vde.' We hypothesize that the prices of these assets are influenced by changes in the stock market, with the relationship being dynamic and responsive to extreme market situations.

In the context of economics mathematics, the Hidden Markov Model (HMM) has been widely utilized to forecast economic regimes and stock prices. In this study, we look at a new HMM approach to stock price prediction. In this work, we show how to measure the HMM’s performance with different numbers of states using the Akaike Information Criterion (AIC), the Hannan– Quinn Information Criterion (HQC), the Bayesian Information Criterion (BIC), and the Bozdogan Consistent Akaike Information Criterion (AIC). After that, we’ll show you how to use HMM to forecast stock prices and how to validate the model with a test set.

Brownian motion is a mathematical description of the random motion of a "large" particle immersed in a fluid and which is not subject to any other interaction than shocks with the "small" molecules of the surrounding fluid. However, this theory is only used to describe the random motion of a large particle. The latter is very useful in finance, for example, the famous Black-Scholes model which evaluates the price of an asset over continuous time. Our report consists of five parts, the first one is reserved for the historical side of the Brownian motion, in the second part we will define mathematically the Brownian motion, as well as its construction in the third part with the help of a random walk or by a Gaussian process. Fourthly, we will see some properties of Brownian motion. And finally the arithmetic and geometric Brownian motion.

This study delves into the analysis of time series data, distinct from cross-sectional data, focusing on the dynamic causal effects and forecasting applications. Using examples such as cigarette consumption in response to tax increases, the study demonstrates the relevance of time series data in estimating and predicting economic phenomena. The discussion covers fundamental concepts, visualization techniques, and the estimation of autoregressive models, emphasizing the importance of stationarity. Empirical applications involve U.S. macroeconomic indicators and financial time series, showcasing the versatility of time series analysis in economic forecasting.

The S&P 500 is a stock market index tracking the performance of 500 large US companies, and it's widely followed. Historical daily prices of the S&P 500 were used for the ARIMA method, and the strategy will be applied. The data can be obtained using "quantmod," The adjusted closing price is chosen to be modelled and predicted as it reflects a stock's actual value by considering dividends, stock splits, and new offerings. The study covers over ten years, from January 4, 2010, to October 13, 2021, and used ^GSPC stock data for analysis.

Data science and sports analytics have opened new doors for machine learning and data mining techniques in basketball performance analysis. This study aims to explore the advanced measures of basketball performance in the National Basketball Association (NBA) using Principal Component Analysis (PCA) and Clustering. PCA identifies the most significant variables while Clustering groups similar observations using unsupervised classification algorithms. In this project, we will analyse NBA player data using PCA and Clustering methods in R to evaluate player performance and help coaches make better decisions. Our results will be presented using graphs and tables to visualise the clusters' impact on the data.

Time series analysis and dynamic modeling are important to study topics with a variety of applications in business, economics, finance, and computer science. The purpose of time series analysis is to examine the projected path of time series observations, build a model to characterize the data structure, and then predict the time series' future values. In finance, exchange rate forecasting is a challenging application of current time series forecasting that is critical to the performance of many businesses and financial organizations. The rates are nonlinear, chaotic, noisy, and non-stationary. On finance-related time series, traditional time series analysis, such as ARIMA models, suffers from low. Because they show a heteroscedasticity behavior. As a result, using Neural Networks (NN) to estimate exchange rates has been proposed as a real alternative. In this study, we will utilize an MLP-NN to predict the next step-ahead exchange rate of EUR/USD. From February 2015 to September 2016, daily data was collected (400 observations). The first 300 must be utilized as training data, while the remaining ones must be used as a testing set.