How GROWTH interacts with LOGEST in an Excel?

[Author]

Hi friends,
I am recently started with the advanced part of an excel. So I don't have much knowledge about the statistical functions that exists in Excel. However, now I am having very little knowledge about the Growth and the Logest. I get confused when I read about the interaction that takes place between the growth and the logest. So thought to ask you guys about it.!! Please explain me how GROWTH interacts with LOGEST in an Excel? It would be much grateful if you provide some examples.!!

[Bigga Lexx]

To explain about the growth function, you will have to create an Excel spreadsheet blank and then need to create the table. You can also get the table which is already made from the Microsoft's site. Once you paste that table into a new Excel worksheet, click Paste Options button and then click Match Destination Formatting. With the pasted range still selected, use one of the following procedures, depending on the version of Excel you are running :

• In Microsoft Office Excel 2007, click the tab, click Format in the Cells group and click Adjust column width.
• In Excel 2003, choose column from the format menu and then click AutoFit Selection.

[Rob Dizzle]

GROWTH and LOGEST can be viewed as interacting in the following steps :

1. Growth is called (known_y, known_x, new_x constant)
2. Call LOGEST growth (known_y's, known_x's, constant, TRUE)
3. We get the regression coefficients from this call to LOGEST. These coefficients appear in the first row of table LOGEST results.
4. For each new_x row, the predicted y-value is calculated using these coefficients and values LOGEST new_x in that row.
5. The value calculated in Step 4 is returned in the appropriate cell growth results corresponding to that new_x row.

[DANIEL 602]

Predictor columns (known values of x) are collinear if at least one column, c, can be expressed as a sum of multiples of others, c1, c2, and other columns. Problems with LINEST in versions of Excel earlier than Excel 2003 are from the collinear predictor columns. Column c is often called a redundant because the information it contains can be constructed from the columns c1, c2, and other columns. The fundamental principle of the existence of collinearity is that results should be not affected by whether a redundant column is included in the original data or remove the original data. Why not look LINEST collinearity in versions of Excel earlier than Excel 2003, this principle was easily violated. Prediction columns are almost collinear if at least one column, c, can be expressed as almost equal to a sum of multiples of others, c1, c2, and other columns. In this case "almost equal" means a very small sum of squared deviations of entries in c from the corresponding entries in the weighted sum of c1, c2, and other columns. "Very little" can be less than 10 ^ (-12), for example.

[Soumen]

In Excel 2003 and later versions of Excel, the message is conveyed not in an alert or a text string, but in the output table LOGEST. Growth has no mechanism to deliver that message to the user. In the results table LOGEST a regression coefficient is one, and whose standard error is zero, corresponding to a ratio of a column that was removed the model. LOGEST Results tables are included in rows 23 to 35 for the output growth of the rows 10 to 16. The entries in cells I24: I25 show a column eliminated redundant predictor. In this case, LOGEST decided to remove the column C (cells K24 coefficients I24, J24, correspond to columns C, B and Excel is constant column, respectively). When there is collinearity present, you can remove any of the columns involved and the choice is essentially arbitrary.

[NIcaBoy]

Collinearity is identified in LOGEST in Excel 2003 and later versions of Excel because LOGEST called LINEST. LINEST uses a completely different approach to solve for the regression coefficients. This approach is QR decomposition. Article LINEST contains a tutorial on the QR decomposition algorithm for a small example. Results are negatively affected growth in versions of Excel prior to Excel 2003 because of inaccurate results in LOGEST which in turn come from inaccurate results in LINEST. LINEST is calculated using an approach had paid no attention to problems of multicollinearity. The existence of collinearity cause rounding errors, inappropriate standard errors of regression coefficients and degrees of freedom inadequate.