Showing 1–7 of 7 results for author: Haddad, R E

Dual-sPLS: a family of Dual Sparse Partial Least Squares regressions for feature selection and prediction with tunable sparsity; evaluation on simulated and near-infrared (NIR) data

Authors: Louna Alsouki, Laurent Duval, Clément Marteau, Rami El Haddad, François Wahl

Abstract: Relating a set of variables X to a response y is crucial in chemometrics. A quantitative prediction objective can be enriched by qualitative data interpretation, for instance by locating the most influential features. When high-dimensional problems arise, dimension reduction techniques can be used. Most notable are projections (e.g. Partial Least Squares or PLS ) or variable selections (e.g. lasso… ▽ More Relating a set of variables X to a response y is crucial in chemometrics. A quantitative prediction objective can be enriched by qualitative data interpretation, for instance by locating the most influential features. When high-dimensional problems arise, dimension reduction techniques can be used. Most notable are projections (e.g. Partial Least Squares or PLS ) or variable selections (e.g. lasso). Sparse partial least squares combine both strategies, by blending variable selection into PLS. The variant presented in this paper, Dual-sPLS, generalizes the classical PLS1 algorithm. It provides balance between accurate prediction and efficient interpretation. It is based on penalizations inspired by classical regression methods (lasso, group lasso, least squares, ridge) and uses the dual norm notion. The resulting sparsity is enforced by an intuitive shrinking ratio parameter. Dual-sPLS favorably compares to similar regression methods, on simulated and real chemical data. Code is provided as an open-source package in R: \url{https://CRAN.R-project.org/package=dual.spls}. △ Less

Submitted 17 January, 2023; originally announced January 2023.

MSC Class: 62J07; 62G08; 62J05 ACM Class: G.3; J.6

Explicit Formula for the n-th Derivative of a Quotient

Authors: Roudy El Haddad

Abstract: Leibniz's rule for the $n$-th derivative of a product is a very well known and extremely useful formula. In this article, we introduce an analogous explicit formula for the $n$-th derivative of a quotient of two functions. Later, we use this formula to derive new partition identities and to develop expressions for some special $n$-th derivatives. Leibniz's rule for the $n$-th derivative of a product is a very well known and extremely useful formula. In this article, we introduce an analogous explicit formula for the $n$-th derivative of a quotient of two functions. Later, we use this formula to derive new partition identities and to develop expressions for some special $n$-th derivatives. △ Less

Submitted 18 July, 2021; v1 submitted 13 April, 2021; originally announced April 2021.

Comments: 11 pages

MSC Class: 26A24 (Primary) 28A15; 11P99 (Secondary)

Journal ref: Open Journal of Mathematical Analysis, Vol. 6 (2022), Issue 2, pp.120-129

Repeated Sums and Binomial Coefficients

Authors: Roudy El Haddad

Abstract: Binomial coefficients have been used for centuries in a variety of fields and have accumulated numerous definitions. In this paper, we introduce a new way of defining binomial coefficients as repeated sums of ones. A multitude of binomial coefficient identities will be shown in order to prove this definition. Using this new definition, we simplify some particular sums such as the repeated Harmonic… ▽ More Binomial coefficients have been used for centuries in a variety of fields and have accumulated numerous definitions. In this paper, we introduce a new way of defining binomial coefficients as repeated sums of ones. A multitude of binomial coefficient identities will be shown in order to prove this definition. Using this new definition, we simplify some particular sums such as the repeated Harmonic sum and the repeated Binomial-Harmonic sum. We derive formulae for simplifying general repeated sums as well as a variant containing binomial coefficients. Additionally, we study the $m$-th difference of a sequence and show how sequences whose $m$-th difference is constant can be related to binomial coefficients. △ Less

Submitted 30 June, 2021; v1 submitted 9 February, 2021; originally announced February 2021.

Comments: 21 pages

MSC Class: 05A10

Journal ref: Open Journal of Discrete Applied Mathematics Vol. 4 (2021), Issue 2, pp. 30-47

Repeated Integration and Explicit Formula for the $n$-th Integral of $x^m (\ln x)^{m'}$

Authors: Roudy El Haddad

Abstract: Repeated integration is a major topic of integral calculus. In this article, we study repeated integration. In particular, we study repeated integrals and recurrent integrals. For each of these integrals, we develop reduction formulae for both the definite as well as indefinite form. These reduction formulae express these repetitive integrals in terms of single integrals. We also derive a generali… ▽ More Repeated integration is a major topic of integral calculus. In this article, we study repeated integration. In particular, we study repeated integrals and recurrent integrals. For each of these integrals, we develop reduction formulae for both the definite as well as indefinite form. These reduction formulae express these repetitive integrals in terms of single integrals. We also derive a generalization of the fundamental theorem of calculus that expresses a definite integral in terms of an indefinite integral for repeated and recurrent integrals. From the recurrent integral formulae, we derive some partition identities. Then we provide an explicit formula for the $n$-th integral of $x^m(\ln x)^{m'}$ in terms of a shifted multiple harmonic star sum. Additionally, we use this integral to derive new expressions for the harmonic sum and repeated harmonic sum. △ Less

Submitted 25 June, 2021; v1 submitted 21 February, 2021; originally announced February 2021.

Comments: 28 pages

MSC Class: 26A36 (Primary) 26A06; 97I50; 11M32; 11P99 (Secondary)

Journal ref: Open Journal of Mathematical Sciences, Vol. 6 (2022), Issue 1, pp. 51-75

Multiple Sums and Partition Identities

Authors: Roudy El Haddad

Abstract: Sums of the form $\sum_{q \leq N_1 < \cdots < N_m \leq n}{a_{(m);N_m}\cdots a_{(2);N_2}a_{(1);N_1}}$ date back to the sixteen century when Viète illustrated that the relation linking the roots and coefficients of a polynomial had this form. In more recent years, such sums have become increasingly used with a diversity of applications. In this paper, we develop formulae to help with manipulating su… ▽ More Sums of the form $\sum_{q \leq N_1 < \cdots < N_m \leq n}{a_{(m);N_m}\cdots a_{(2);N_2}a_{(1);N_1}}$ date back to the sixteen century when Viète illustrated that the relation linking the roots and coefficients of a polynomial had this form. In more recent years, such sums have become increasingly used with a diversity of applications. In this paper, we develop formulae to help with manipulating such sums (which we will refer to as multiple sums). We develop variation formulae that express the variation of multiple sums in terms of lower order multiple sums. Additionally, we derive a set of partition identities that we use to prove a reduction theorem that expresses multiple sums as a combination of simple (non-recurrent) sums. We present a variety of applications including applications concerning polynomials and MZVs as well as a generalization of the binomial theorem. Finally, we establish the connection between multiple sums and a type of sums called recurrent sums. By exploiting this connection, we provide additional partition identities for odd and even partitions. △ Less

Submitted 25 January, 2021; originally announced February 2021.

Comments: 36 pages. arXiv admin note: substantial text overlap with arXiv:2101.09089

MSC Class: 11P84 (Primary) 11B73; 11M32; 11C08 (Secondary)

Journal ref: Notes on Number Theory and Discrete Mathematics, 28(2), 200-233 (2022)

Recurrent Sums and Partition Identities

Authors: Roudy El Haddad

Abstract: Sums of the form $\sum_{N_m=q}^{n}{\cdots \sum_{N_1=q}^{N_2}{a_{(m);N_m}\cdots a_{(1);N_1}}}$ where the $a_{(k);N_k}$'s are same or distinct sequences appear quite often in mathematics. We will refer to them as recurrent sums. In this paper, we introduce a variety of formulas to help manipulate and work with this type of sums. We begin by developing variation formulas that allow the variation of a… ▽ More Sums of the form $\sum_{N_m=q}^{n}{\cdots \sum_{N_1=q}^{N_2}{a_{(m);N_m}\cdots a_{(1);N_1}}}$ where the $a_{(k);N_k}$'s are same or distinct sequences appear quite often in mathematics. We will refer to them as recurrent sums. In this paper, we introduce a variety of formulas to help manipulate and work with this type of sums. We begin by developing variation formulas that allow the variation of a recurrent sum of order $m$ to be expressed in terms of lower order recurrent sums. We then proceed to derive theorems (which we will call inversion formulas) which show how to interchange the order of summation in a multitude of ways. Later, we introduce a set of new partition identities in order to then prove a reduction theorem which permits the expression of a recurrent sum in terms of a combination of non-recurrent sums. Finally, we apply this reduction theorem to a recurrent form of two famous types of sums: The $p$-series and the sum of powers. △ Less

Submitted 22 January, 2021; originally announced January 2021.

Comments: 34 pages

MSC Class: 11P84 (Primary) 11B73; 11M32; 05A18 (Secondary)

Journal ref: Notes on Number Theory and Discrete Mathematics, 28(2), 167-199 (2022)

Stratified Monte Carlo simulation of Markov chains

Authors: Rana Fakhereddine, Rami El Haddad, Christian Lécot

Abstract: We present several Monte Carlo strategies for simulating discrete-time Markov chains with continuous multi-dimensional state space; we focus on stratified techniques. We first analyze the variance of the calculation of the measure of a domain included in the unit hypercube, when stratified samples are used. We then show that each step of the simulation of a Markov chain can be reduced to the numer… ▽ More We present several Monte Carlo strategies for simulating discrete-time Markov chains with continuous multi-dimensional state space; we focus on stratified techniques. We first analyze the variance of the calculation of the measure of a domain included in the unit hypercube, when stratified samples are used. We then show that each step of the simulation of a Markov chain can be reduced to the numerical integration of the indicator function of a subdomain of the unit hypercube. Our approach for Markov chains simulates N copies of the chain in parallel using stratified sampling and the copies are sorted after each step, according to their successive coordinates. We analyze variance reduction on examples of pricing of European and Asian options: enhanced efficiency of stratified strategies is shown. △ Less

Submitted 21 March, 2016; originally announced March 2016.

Comments: 16 pages, 4 figures