# Asim Srivastava's  Instablog

Asim Srivastava
Send Message
I'm a final year undergraduate student at the University of Warwick and I have interests in Quantitative Financial Modeling, computer programming, pure and applied mathematics, statistics and financial trading. I have been working on a quant model which provides forecasts for stock prices on a... More
My blog:
Quantitative Forecasting of Stock Prices
• ##### Quant Forecasting Of Stock Prices 2 comments
Feb 13, 2013 2:31 AM | about stocks: GOOG, AAPL, AMZN, PCLN, IBM

Introduction

Hello Everyone,

I have recently developed a quant model which uses a form of Fourier Analysis in order to forecast stock prices and I am currently testing it on live market data.

Based on real time day trading on a virtual portfolio, I have been able to achieve consistent daily returns of 0.1% (annualized return of 25% for 250 trading days) on a portfolio consisting of a variety of stocks.

I will be publishing my forecasts regularly and you can follow them to see how accurate they are.

Regards and enjoy,

Asim

Check out the full blog at: quantforecasting.blogspot.co.uk

First Forecasts for 13/02/2013 at 03:03 GMT

Name Direction Confidence Fit Robust Forecast OK?
GOOG Down 30% Bad Yes NO
AAPL Down 50% Ok Yes Check
AMZN UP 41% Ok Yes Check
IBM UP 54% OK Yes Yes
PCLN UP 83% OK Yes Yes

Here are a few things you should know:

Direction: This tells you what the general direction of movement of the stock will be today and not necessarily where it will close relative to yesterday's price.

Confidence: This is the most important diagnostic measure for these forecasts. It basically tells you how likely the forecast is to be correct, or simply, the confidence I have in the forecast. A high confidence generally means the forecast is likely to be true. The value for confidence is calculated from the model itself.

Fit: A simple diagnostic check which ensures the model provides a good statistical fit to the data already observed. The fit may be Good, Ok or Bad, indicating the model fits the data very well, just alright, or not fit at all, respectively.

Robust: This check tells you whether or not the forecast is robust. Meaning, does changing the model parameters slightly drastically alter the results? A robust result is one which is invariant under small changes in parameter values. Basically, a robust forecast is a good thing. A non robust forecast simply means it may very likely be wrong.

Forecast OK?: This summarizes whether or not the forecast is alright overall, taking into account all the diagnostic information. 'Yes' and 'No' are obvious, if all the diagnostic measures are ok, i.e. Confidence levels are high, the Fit is good and the forecast is Robust, then it means that overall, forecast is alright. If all or most diagnostic measures were poor, then it would mean that the forecast is most likely incorrect. 'Check' means its not entirely obvious from the diagnostic information whether or not the forecast is ok and so no meaningful conclusion can be reached.

Hopefully that covers just about everything, if there's anything else you'd like to know, feel free to ask!

***Disclaimer***:

All information provided herein and in all further posts is to be used for educational purposes only and not to be considered as advice or recommendation of any sort. I cannot guarantee the accuracy of the forecasts I put up as the method I am using is still under development and testing. Consequently you should not base your investment decisions on these forecasts, and I will not be held liable for any losses incurred as a result of using these forecasts.

Stocks: GOOG, AAPL, AMZN, PCLN, IBM
Back To Asim Srivastava's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

• Asim Srivastava
, contributor

Author’s reply » Returns for 13/02/2013:

AAPL - Shorted: 468.25, Covered: 466.21, Return: 0.435%
AMZN - Bought: 258.07, Sold: 260.55, Return: 0.96%
PCLN - Bought: 703.56, Sold: 705.20, Return: 0.233%

Total Amount Invested: 450219
Today's Profit: 1520
Available Capital: 1,000,000

Daily Return: 0.152%
14 Feb 2013, 05:59 PM Reply Like
• signalsprovider
, contributor
Comment (1) | Send Message

The process of attempting to predict the future condition of the economy. This involves the use of statistical models utilizing variables sometimes called indicators. Some of the most well-known economic indicators include inflation and interest rates, GDP growth/decline, retail sales and unemployment rates.
stock forecast
27 Jan 2014, 01:50 AM Reply Like