Artificial Intelligence

 Association Rules |Clustering |Decision Trees| Naive Bayes| Regression Analysis| Neural Networks| Time Series |Text Mining |

Genetic Programming

Methods for time series analyses are often divided into two classes: frequency-domain methods and time-domain methods. The former centre around spectral analysis and recently wavelet analysis, and can be regarded as model-free analyses well-suited to exploratory investigations. Time-domain methods have a model-free subset consisting of the examination of auto-correlation and cross-correlation analysis, but it is here that partly and fully-specified time series models make their appearance.

Time Series is most effective Algorithm to predict Future Price, Sales trend and fairly accurate for next few days. Predictions cannot track very sharp price movements but just after 2 or 3 days this algorithm again corrects and tracks with uncanny accuracy. Sometimes the closing price predicted is accurate to second decimal place. Recom Systems has over 3 years predictions for all the Indian stocks on the NSE and our results of predicted values for next 3 to 5 days were with in 2% accuracy. For Sales Forecast similar predictions are achievable.

Its fields of application range from neurophysiology to astrophysics and it covers such well-known areas as economic forecasting, study of biological data, control systems, signal processing and communications and vibrations engineering.

Time Series performs univariate and multivariate analysis and enables you to explore both stationary and non-stationary models. Standard methods such as Yule-Walker, Levinson-Durbin, long auto-regression, Hannan-Rissanen are available.

Time series data is very important for correct forecasting, this data requires proper conditioning and filtration of the erroneous data. While data is fed through various channels errors crop up and these require elimination before data is feed to the time series algorithms.  Recom's experience in Stock Market prediction of stocks and indices have been that a accuracies of less than 1% are achievable with properly conditioned data. many times data is not available in the desired format and this require a change in the format before prediction.

Time series studies of environmental exposures often involve comparing daily changes in a toxicant measured at a point in space with daily changes in an aggregate measure of health. Spatial misalignment of the exposure and response variables can bias the estimation of health risk and the magnitude of this bias depends on the spatial variation of the exposure of interest. In air pollution epidemiology, there is an increasing focus on estimating the health effects of the chemical components. A sample of time series chart is shown below: