By Niu Shu-fen, Wang Guo-xin, Sun Xiao-ling
During this paper, a brand new branch-and-bound set of rules in accordance with the Lagrangian twin rest and non-stop rest is proposed for discrete multi-factor portfolio choice version with roundlot limit in monetary optimization. This discrete portfolio version is of integer quadratic programming difficulties. The separable constitution of the version is investigated through the use of Lagrangian leisure and twin seek. Computational effects exhibit that the set of rules is in a position to fixing real-world portfolio issues of info from US inventory marketplace and randomly generated try out issues of as much as one hundred twenty securities.
Read or Download A branch-and-bound algorithm for discrete multi-factor portfolio optimization model PDF
Best algorithms and data structures books
The layout of Innovation illustrates the right way to layout and enforce efficient genetic algorithms-genetic algorithms that clear up demanding difficulties fast, reliably, and accurately-and how the discovery of powerfuble genetic algorithms quantities to the production of an efficient computational thought of human innovation.
Fresh years have witnessed a dramatic raise of curiosity in subtle string matching difficulties, specifically in info retrieval and computational biology. This booklet provides a pragmatic method of string matching difficulties, targeting the algorithms and implementations that practice top in perform.
The Lewis proposal of acids and bases is mentioned in each common, natural and inorganic chemistry textbook. this can be frequently only a descriptive remedy, because it isn't really attainable to plan a unmarried numerical scale compatible for all events. in spite of the fact that quantitative Lewis acid-base chemistry may be built by way of compiling reaction-specific basicity scales which are utilized in particular branches of chemistry and biochemistry.
- Understanding Microsoft Virtualization Solutions: From the desktop to the datacenter
- Computing for Scientists and Engineers: A Workbook of Analysis, Numerics, and Applications
- Genome Exploitation: Data Mining the Genome (Stadler Genetics Symposia Series)
- The Little Data Book on Information and Communication Technology 2008 (Little Data Book on Information and Communication Technology)
Extra info for A branch-and-bound algorithm for discrete multi-factor portfolio optimization model
In contrast, other schemes such as Gray coding reduce this problem. In Gray coding, the object is to create a code such that a single integer change only requires a 1-bit change in the binary genotype. This means that adjacent solutions in the (integer or real-valued) search space will be adjacent in the (binary) encoding space as well, requiring fewer mutations to discover. The Gray coding rule starts with a string of all zeros for the integer value zero, and to create each subsequent integer in sequence the rule successively ﬂips the right-most bit that produces a new string.
027451. could be converted into a real value as follows: 0 + 255 Although the above decoding scheme for a binary string is quite simple, it can suﬀer from Hamming cliﬀs, in that sometimes a large change in the 44 3 Evolutionary Methodologies genotype is required to produce a small change in the resulting integer value. 5, it can be seen that the underlying genotype needs to change in all three bit positions. These Hamming cliﬀs can potentially create barriers that the GA could ﬁnd diﬃculty in passing.
N , each of which encode a solution, is randomly initialised and evaluated using a ﬁtness function f . During the search process, each individual (j) is iteratively reﬁned. The modiﬁcation process has three steps: i. Create a variant vector which encodes a solution, using randomly selected members of the population (mutation step). 50 3 Evolutionary Methodologies ii. Create a trial vector, by combining the variant vector with j (crossover step). iii. Perform a selection process to determine whether the newly-created trial vector replaces j in the population.
A branch-and-bound algorithm for discrete multi-factor portfolio optimization model by Niu Shu-fen, Wang Guo-xin, Sun Xiao-ling