https://github.com/didierrlopes/univariatetimeseriesforecast
PhD Thesis: "Data Science in the Modeling and Forecasting of Financial Timeseries: from Classic methodologies to Deep Learning"
https://github.com/didierrlopes/univariatetimeseriesforecast
arima-model deep-learning lstm-neural-networks phd-thesis time-series time-series-forecasting
Last synced: 11 months ago
JSON representation
PhD Thesis: "Data Science in the Modeling and Forecasting of Financial Timeseries: from Classic methodologies to Deep Learning"
- Host: GitHub
- URL: https://github.com/didierrlopes/univariatetimeseriesforecast
- Owner: DidierRLopes
- Created: 2020-01-27T20:56:17.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2021-07-01T21:30:24.000Z (about 5 years ago)
- Last Synced: 2025-07-17T04:07:20.915Z (12 months ago)
- Topics: arima-model, deep-learning, lstm-neural-networks, phd-thesis, time-series, time-series-forecasting
- Language: Jupyter Notebook
- Homepage:
- Size: 18.1 MB
- Stars: 33
- Watchers: 2
- Forks: 3
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Univariate Time Series Forecast
This study was developed with Filipe Roberto Ramos (https://ciencia.iscte-iul.pt/authors/filipe-roberto-de-jesus-ramos/cv) for his phD thesis entitled "Data Science in the Modeling and Forecasting of Financial Timeseries: from Classic methodologies to Deep Learning". Submitted in 2021 to Instituto Universitário de Lisboa - ISCTE Business School, Lisboa, Portugal.
## Notebooks
### ExploratoryDataAnalysis
* Imports and Defines
* Data Inspection
* Data Visualization
* Data Analysis
* Hypothesis Test
### ARIMA and SARIMA
* Imports and Defines
* Imports
* Define Functions
* Define Univariate Time-Series
* Stationarity of the Time-Series
* Data transformation and its graphical representation
* Normality tests
* Jarques-Bera
* Kolmogorov-Smirnov
* Unit Root and Stationarity Tests
* The Augmented Dickey-Fuller test
* Kwiatkowski-Phillips-Schmidt-Shin
* Correlation plots
* (S)ARIMA Selection
* Pre-processing
* Model training
* Model Comparison based on Information Criteria
* Selected Models Information Criteria Comparison
* Selected Models Cross-Validation
* Model Validation
* Model Residual Analysis
* Normality tests
* Kurtosis and Kurtosis Test
* Skew and Skewness Test
* Jarque-Bera and Kolmogorov-Smirnov tests
* Engle's Test for Autoregressive Conditional Heteroscedasticity (ARCH)
* Test for No Autocorrelation
* Brock–Dechert–Scheinkman test
* Breusch-Godfrey test [NOT IN SARIMA]
* Box-Pierce and Ljung-Box tests
* QQplot
* Auto-correlation and Partial Auto-correlation functions
* Model Prediction
* Model Prediction Overview
### ExponenTialSmoothing
* Imports and Defines
* Imports
* Define Functions
* Define Univariate Time-Series
* Data Visualization
* Model Training
* Single Exponential Smoothing
* TS (N, N) - Simple Exponential Smoothing
* Double Exponential Smoothing
* TS (A, N) - Holt’s linear method
* TS (Ad, N) - Additive damped trend method
* Triple Exponential Smoothing
* TS (N, A) method
* TS (A, A) - Additive Holt-Winters method
* TS (Ad, A) method
* TS (N, M) method
* TS (A, M) Multiplicative Holt-Winters’ method
* TS (Ad, M) Holt-Winters’ damped method
* Model Selection
* Model Validation
* Normality Test
* Kurtosis and Kurtosis Test
* Skew and Skewness Test
* Jarque-Bera test
* Kolmogorov-Smirnov test
* Engle's Test for Autoregressive Conditional Heteroscedasticity (ARCH)
* Test for No Autocorrelation
* Brock–Dechert–Scheinkman test
* Box-Pierce and Ljung-Box tests
* QQplot
* Plot Auto-correlation and Partial Auto-correlation functions
* Model Prediction
* Model Prediction Overview
### DeepNeuralNetwork
And **DNN_OurApproach**
* Imports and Defines
* Imports
* Define Functions
* Define Univariate Time-Series
* Training Deep Neural Network
* Data Pre-Processing
* Visualization of Data Pre-Processed
* Model Selection (tune hyper-parameters)
* Cross-Validation
* Compute Cross-Validation Errors
* Cross-Validation Performance
* Cross-Validation Plot
* Model forecasting
* Perform statistics on predictions
* Statistics cleaning
* Plot Prediction
* Plot Forecast in-sample vs out-sample
## Citation
APA
`Ramos, F. (2021). Data Science na Modelação e Previsão de Séries Económico-financeiras: das Metodologias Clássicas ao Deep Learning. (PhD Thesis submitted, Instituto Universitário de Lisboa - ISCTE Business School, Lisboa, Portugal).`
```
@phdthesis{FRRamos2021,
AUTHOR = {Filipe R. Ramos},
TITLE = {Data Science na Modelação e Previsão de Séries Económico-financeiras: das Metodologias Clássicas ao Deep Learning},
PUBLISHER = {PhD Thesis submitted, Instituto Universitário de Lisboa - ISCTE Business School, Lisboa, Portugal},
YEAR = {2021}
}
```