Forecasting Methods in Various Applications Using Algorithm of Estimation Regression Models and Converting Data Sets into Markov Model

Table 3

Regression models polynomial of the first, second, and third degree.

Model	Depth	Regression equation
y = ax^3 + bx^2 + cx + d	1.34 ≤ x ≤ 2.5	Temp (2) = 9.5611x^3 − 55.367x^2 + 104.34x − 43.508
		pH (2) = −3.854x^3 + 23.964x^2 − 47.844x + 38.775
		ORP (2) = −519.66x^3 + 3246.5x^2 − 6421.5x + 3872.8
		DO (2) = −38.682x^3 + 238.37x^2 − 473.74x + 307.87
		RES (2) = −9.2607x^3 + 31.564x^2 − 14.813x + 2.4553
		SAL (2) = −20.088x^3 + 155.97x^2 − 374.06x + 313.21
		SSG (2) = −18.391x^3 + 137.34x^2 − 321.19x + 260.33
		TN (2) = −1226.9x^3 + 6963.8x^2 − 12820x + 7762.6
		cd (2) = −0.0467x^3 + 0.2661x^2 − 0.4855x + 0.3171
		cu (2) = −0.041x^3 + 0.242x^2 − 0.4578x + 0.3485
		zn (2) = 0.0325x^3 − 0.2051x^2 + 0.4232x − 0.2459
		pb (2) = 0.0038x^3 − 0.025x^2 + 0.054x − 0.0307
		fe (2) = −4.2495x^3 + 23.521x^2 − 41.898x + 24.608
	0.38 < x ≤ 2.5	Total pb (2) = 0.006x^3 + 0.076x^2 − 0.3468x + 0.3533
y = ax + b&y = ax^2 + bx + c	0.13 ≤ x < 0.3	Total pb (2) = −0.4375x + 0.1359
	0.13 ≤ x < 0.3	EC (2) = −63275x + 71243
	0.3 ≤ x < 0.38	EC (2) = −2E + 08x^2 + 1E + 08x − 2E + 07
	0.38 ≤ x < 2	EC (2) = −62562x^2 + 10838x + 31641
	2 ≤ x ≤ 2.5	EC (2) = 29094x − 14309
	0.13 ≤ x < 0.3	COD (2) = 13281x − 1251.6
	0.3 ≤ x < 0.38	COD (2) = 9E + 06x^2 − 5E + 06x + 83615
	0.38 ≤ x < 2	COD (2) = 12169x − 16183
	2 ≤ x ≤ 2.5	COD (2) = 550x + 50
	0.13 ≤  x < 0.3	TDS (2) = −40494x + 45594
	0.3 ≤  x < 0.38	TDS (2) = −3E + 07x^2 + 2E + 07x − 3E + 06
	0.38 ≤  x < 2	TDS (2) = −40036x^2 + 69359x + 20251
	2 ≤  x ≤  2.5	TDS (2) = 18620x − 9158