𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Comment on the evaluation of the arrhenius parameters by the least squares method

✍ Scribed by R. J. Cvetanovic; D. L. Singleton


Publisher
John Wiley and Sons
Year
1977
Tongue
English
Weight
450 KB
Volume
9
Category
Article
ISSN
0538-8066

No coin nor oath required. For personal study only.

✦ Synopsis


Abstract

Arrhenius parameters are frequently evaluated incorrectly by applying the least squares method to the logarithmic form of the Arrhenius equation without simultaneously transforming the statistical weights as required for the change of variable from k to ln k. This has been mentioned briefly in an earlier paper. In the present communication the correct procedures are discussed and illustrated by several examples of evaluations. In particular, Arrhenius parameters calculated by the Taylor series expansion of the exponential form of the Arrhenius equation are compared with those calculated from the logarithmic form, using an exact and two approximate transformations of the statistical weights. The comparisons indicate thatthe preferred procedure for obtaining Arrhenius parameters is either the Taylor series method or the logarithmic method with proper transformation of the experimentally determined statistical weights of the rate constants k~i~. The common approximation of assuming equal statistical weights of ln k~i~ when the logarithmic form of the Arrhenius expression is used is shown not to be always appropriate, and reasons forthis are given.


πŸ“œ SIMILAR VOLUMES


Molecular structure by the constrained d
✍ S. Subhasini; K. Sundaram πŸ“‚ Article πŸ“… 1983 πŸ› Wiley (John Wiley & Sons) 🌐 English βš– 486 KB πŸ‘ 2 views

## Abstract A method for simulating three‐dimensional molecular structure is presented. The method can satisfy any number of constraints on the molecular structure. The constraints can pertain to the constancy of individual distances and angles or be related to point‐group symmetry. A simulated thr

On the accuracy of total least squares a
✍ Sabine Van Huffel; Joos Vandewalle πŸ“‚ Article πŸ“… 1989 πŸ› Elsevier Science 🌐 English βš– 388 KB

Abstroct~Every linear parameter estimation problem gives rise to an overdetermined set of linear equations AX ~ B which is usually solved with the ordinary least squares (LS) method. Often, both A and B are inaccurate. For these cases, a more general fitting technique, called total least squares (TL