I am trying to forecast prices with exponential moving average method. The equation for EMA = [(Closing * k) + (EMA( y ) * (1-k)] where: Closing is closing price of today, k is the weighted multiplier, EMA( y ) is the previous EMA, and 1 - k is 1 - weighted multipler. This gives the EMA for today. I would like to know how can I modify this equation to forecast the EMA of the next day?
How to forecast next day Exponential Moving Average
- Thread starter mahmoud899
- Start date
- Watchers 7
Surely you would need to know tomorrow's closing price in order to forecast tomorrow's EMA?
And if you could know tomorrow's price, why would you want to forecast EMA?
And if you could know tomorrow's price, why would you want to forecast EMA?
The Exponential Moving Average equation requires the closing price of today to be able to be calculated. I am trying to forecast the price of tomorrow using the Exponential Moving Average, maybe you didn't understand my question, is there a way to adjust the equation or is there another equation which gives you the Exponential Moving Average forecast of tomorrow?
This can be done with Bernoulli's equation neglecting friction. Its far more complicated including friction as that would take us into Partial Differential Equations - beyond the scope here
PT = PS +1/2 µv²
See https://apps.dtic.mil/dtic/tr/fulltext/u2/a075515.pdf (includes source code for SUBROUTINE LNPRED).
I wouldn't call it the Holy Grail, but linear prediction works reasonably well when the data is detrended.
No problem, take values of exponential moving average for one year then simulate ARIMA model using this time series. If the model you get is stationary you can make forecasts. But unfortunately you are likely to get random walk model because underlying time series - i.e. closing prices are most likely AR(1) model
Have you tried just shifting the EMA 1-bar forward?
OR,
Given that the only unknown variable you seem to show is the "Closing", could you work out the average move (velocity) of closing prices are, and project forward?
EG, if the last 3 Closing prices are 100, 110, 118, the market has closed (118 - 100) = 18 points higher per past 3 days, which is a velocity of 6 per day. Could you project the Closing as 124 (118 + 6) and go from there?
Predictions imply the market wont spike, surge, stall, congest, etc. Past performance is no guarantee of (you know the rest)
Some people use trendlines.
Good luck with this.