Charts represent S&P500 forecast for June 1-5, 2009. The calculations* have been performed using pattern similarity prediction (price and volume) and neural network tool.
Composite prediction in numbers –
1-3: fluctuation within 900-920
4-5: above 920
Charts represent S&P500 forecast for May 26-29, 2009. The calculations* have been performed using pattern similarity prediction (price and volume) and cycle analysis.
Major points (composite prediction) in numbers:
27 – below 880;
29 – above 910;
Charts represent S&P500 forecast for May 18-22, 2009. The calculations* have been performed using pattern similarity prediction (price and volume) and Neural Network (technical indicators as inputs).
Major points (composite prediction) in numbers:
18 – above 884;
21 – below 870;
It seems now economical and technical analysis data make investors less optimistic**. Since the trend is that more and more factors start suggesting a downturn in the short term, it may happen soon.
**) source – May 2009 Stock Investing News
Table shows ^GSPC closing price on each trading day from May 5 to May 15. Results calculated using experimental tool NNSPP-2 (Neural Network). Data presented for testing purpose only.
Comparative Analysis (added on 2009-05-15):
date actual predicted difference
2009-05-05 903.80 904.17 0.37 0.04%
2009-05-06 919.53 905.04 14.49 1.58%
2009-05-07 907.39 915.63 8.24 0.91%
2009-05-08 929.23 910.45 18.78 2.02%
2009-05-11 909.24 906.58 2.66 0.29%
2009-05-12 908.35 912.47 4.12 0.45%
2009-05-13 883.92 915.30 31.38 3.55%
2009-05-14 893.07 913.80 20.73 2.32%
2009-05-15 882.88 918.73 35.85 4.06%
------
average: 1.69%
^GSPC forecast for May-July using Neural Network
This is an approximate ^GSPC forecast for May-July (using Neural Network from experimental tool NNSTP-2).
Major points in numbers:
Reminder: Forecast built on statistical methods may fail because of redistribution of driving factors – news, investors’ sentiments, regulatory impact, etc.
Important to notice: The recent forecast by pattern similarity shows earlier selloff.
^GSPC for May-June 2009, predicted by pattern similarity
The algorithm of InvAn prediction using pattern similarity is the following. It searches for the best match from the internal database by scanning all historical data. The ranking of all possible matches is calculated on the basis of minimum deviation and maximum correlation within given historical period (number of candlesticks). Pattern matching is performed using open, high, low, and close prices and volume data. When scanning is completed, it composes forecast using several best matched patterns (top ranked). The composite result is built as a weighted average with weights proportionally patterns’ ranks. More details can be found at www.addaptron.com
Major points in numbers: