MySql: database values normalization

gistfile1.sql

/**
* Database values normalization shows diferences betwen @OUTPUT_LOW - @OUTPUT_HIGH
* its sample for forex database with tips table and 5 colls
* open FLOAT(8,5), high FLOAT(8,5), low FLOAT(8,5), close FLOAT(8,5),
* volume TINYINT(4), symbol (ENUM), time (DATETIME).
* tested on MySql
*
* @author AgBorkowski [email protected]
* @uses http://www.heatonresearch.com/content/really-simple-introduction-normalization (java)
* @link http://blog.aeonmedia.eu/2011/02/mysql-database-values-normalization
*/

/**
* Important params for charset equal with your table encoding
*/
SET NAMES 'utf8' COLLATE 'utf8_unicode_ci';

/**
* The low range of the input (un-normalized data)
*/
SET @INPUT_LOW := 0;

/**
* The high range of the input (un-normalized data)
*/
SET @INPUT_HIGH := 0;

/**
* The low range of the output (normalized data)
*/
SET @OUTPUT_LOW := 0;

/**
* The high range of the output (normalized data)
*/
SET @OUTPUT_HIGH := 1;

/**
* The symbol of stock
*/
SET @SYMBOL := 'EURUSD';


/**
* Set @INPUT_LOW and @INPUT_HIGH
*/
SELECT @INPUT_LOW := MIN(`low`), @INPUT_HIGH := MAX(`high`)
FROM forex.tips
WHERE symbol = @SYMBOL
GROUP BY symbol
SELECT
((`open` - @INPUT_LOW) / (@INPUT_HIGH - @INPUT_LOW)) * (@OUTPUT_HIGH - @OUTPUT_LOW) + @OUTPUT_LOW AS `open`,
((`high` - @INPUT_LOW) / (@INPUT_HIGH - @INPUT_LOW)) *(@OUTPUT_HIGH - @OUTPUT_LOW) + @OUTPUT_LOW AS `high`,
((`low` - @INPUT_LOW) / (@INPUT_HIGH - @INPUT_LOW)) * (@OUTPUT_HIGH - @OUTPUT_LOW) + @OUTPUT_LOW AS `low`,
((`close` - @INPUT_LOW) / (@INPUT_HIGH - @INPUT_LOW)) * (@OUTPUT_HIGH - @OUTPUT_LOW)+ @OUTPUT_LOW AS `close`,
volume
FROM forex.tips
WHERE symbol = @SYMBOL;