Embedded Files
(Gerasymov, 2024, n.p.)
Time series forecasting
Time series forecasting
As technology advances, traders increasingly explore AI and ML, gradually mastering these tools to enhance their strategies. A key application is using AI/ML models to forecast stock prices by analyzing historical data, identifying trends, and predicting future values. This is known as Time Series Forecasting.
This involves training a model to predict what's going to happen based on the prehistoric data.
(Gerasymov, 2024, n.p.)
Simple Exponential Smoothing (SES)
Simple Exponential Smoothing (SES)
(Gerasymov, 2024, n.p.) says, "This technique prioritizes recent data by applying a smoothing constant, making it particularly effective in volatile markets where new trends emerge rapidly."
Key Points:
Cannot automatically account for trends or seasonality.
Advanced variations like damped trend models incorporate these crucial factors into predictions.
(Benjamins, 2017, p. 42)
Holt-Winter's Exponential Smoothing (HWES)
Holt-Winter's Exponential Smoothing (HWES)
Another advanced approach is Holt-Winter’s Exponential Smoothing (HWES), also known as Triple Exponential Smoothing.
Key Points:
(Jauhari, 2023, n.p.) says, "HWES builds upon Holt’s method by adding a seasonal component, making it capable of handling data with trends, seasonality, and noise. This complexity makes HWES a powerful model for time series forecasting in trading."
(Gerasymov, 2024, n.p.)
ARIMA Model
ARIMA Model
ARIMA is used for time-series analysis, focusing on extracting meaningful characteristics and data from past data to predict future stock prices. (Dhyani et al., 2020)
Key Points:
Helps understand past trends to predict future behavior.
Supporting Perspective
Ardakani and Saenz (2023)(p. 27) gave the Idea, AI technologies are reshaping the trading landscape by improving operational efficiency and reducing human dependencies, making them indispensable for modern markets. This highlights how AI-driven automation boosts productivity by analyzing vast amounts of data, enabling faster and more accurate decision-making. This perspective reinforces the transformative potential of AI in trading, reducing human involvement and allowing systems to quickly adapt to market changes.
Contrasting Perspective
On the other hand, Huang et al. (2019) highlight, The success of AI in HFT hinges on access to high-performance hardware and ultra-low latency networks. Firms lacking these resources risk falling behind. Which implies that while AI systems promise efficiency, the lack of high-performance hardware like a good CPU or monitor can be a limiting factor for smaller firms. Huang et al. (2019) argue that the success of automated trading heavily depends on technology and infrastructure, like high performance hardware. They suggest that while AI can be powerful, it has practical limitations.
Page updated
Google Sites
Report abuse