Using ADX, Moving Averages, and Time-Series Models to Estimate Trends

Summary

The document surveys several ways to assess whether a stock is trending. It identifies the Average Directional Index (ADX) as a technical indicator designed to gauge trend strength and mentions moving averages and moving-average crossovers as additional tools. These methods provide structured alternatives to drawing trend lines by eye, though the text does not compare their settings or performance.

It also points to time-series forecasting approaches, including ARIMA models and exponential smoothing, as methods that can estimate trend in observed data. The responses warn that estimating a trend is not the same as reliably forecasting stock prices: prices may be difficult to predict even when a trend measure can be calculated. No empirical tests, asset-specific guidance, or signal thresholds are supplied. The discussion is therefore an introductory list of techniques, not evidence that any one indicator or model will produce profitable trading signals.

Key ideas

• ADX can be used to assess whether a security is trending and to gauge trend strength.
• Moving averages and their crossovers are additional technical approaches to trend estimation.
• ARIMA and exponential smoothing models can represent trend in time-series data.
• A measurable trend does not imply that future stock prices are reliably predictable.

Tags

# Trend estimation techniques


# Trend estimation techniques












What is the best way (most common) to discover if a stock is trending or not? (Despite drawing a line). A hurst exponent? A linear regression maybe?

## Answer by Chris Degnen (score 1, accepted)

https://quant.stackexchange.com/a/11341

See the average directional movement index (ADX):-

> "At its most basic the Average Directional Index (ADX) can be used to determine if a security is trending or not."

http://stockcharts.com/school/doku.php?id=chart_school%3atechnical_indicators%3aaverage_directional_#trend_strength

It's easy to calculate. There is an example spreadsheet on the link page.

## Answer by Richi Wa (score 2)

https://quant.stackexchange.com/a/11363

As in the answer by Chris Degen, various technical indicators for trend estimation exist (just to mention one additional: the moving average and moving average cross over).

In this free ebook by Rob Hyndman and George Athana­sopou­los present various time series techniques for forecasting. These are ARIMA models and exponential smoothing methods. They can be used to estimate a trend in time series.

But be careful if I remember correctly then they state that stock prices are hard to forecast (if not impossible).

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.