Geomechanical model and sanding onset assessment: A field case study in Vietnam

Sand production is a key issue when selecting and applying completion solutions like open holes, screens or perforated liners. This problem can be seen in several types of reservoirs such as weakly consolidated and non-consolidated carbonates. The paper presents a method to model wellbore failures for sanding prediction. Our study shows that the potential sand risk in this field is defined by the rock strength rather than the in-situ stress. If the rock is sufficiently competent, the potential of sand production is negligible, and the development wells can be completed conventionally without any downhole sand control for the reservoir pressure above 1,280 psi and the maximum drawdown pressure of 2,380 psi.

Maximum Drawdown, Recovery, and Momentum

We empirically test predictability on asset price using stock selection rules based on maximum drawdown and its consecutive recovery. In various equity markets, monthly momentum- and weekly contrarian-style portfolios constructed from these alternative selection criteria are superior not only in forecasting directions of asset prices but also in capturing cross-sectional return differentials. In monthly periods, the alternative portfolios ranked by maximum drawdown measures exhibit outperformance over other alternative momentum portfolios including traditional cumulative return-based momentum portfolios. In weekly time scales, recovery-related stock selection rules are the best ranking criteria for detecting mean-reversion. For the alternative portfolios and their ranking baskets, improved risk profiles in various reward-risk measures also imply more consistent prediction on the direction of assets in future. Moreover, turnover rates of these momentum/contrarian portfolios are also reduced with respect to the benchmark portfolios. In the Carhart four-factor analysis, higher factor-neutral intercepts for the alternative strategies are another evidence for the robust prediction by the alternative stock selection rules.

A Genetic Algorithm Hyper-parameter Optimization of Ensemble Approach: Strategi Prediksi Saham Mempertimbangkan Indikator Teknikal & Sentimen Berita

Investor harus memprediksi saham dengan tepat agar keuntungan maksimal sekaligus terhindar kebangkrutan. Namun bursa saham sulit dideteksi situasinya. Perilakunya berubah-ubah dipengaruhi berbagai faktor seperti situasi politik, ekonomi perusahaan dan global, maupun ekspektasi investor yang tersedia melalui berita. Penelitian ini bertujuan mengembangkan model yang dapat memprediksi saham lebih akurat mengkombinasikan indikator teknikal saham dan sentimen berita. Genetic algorithm (GA) mengoptimalisasi beberapa ensemble decision tree-based yang ditumpuk menggunakan metode stacked-generalization dengan konsep meta-learner digunakan dalam penelitian ini. Terdapat lima tahapan utama metodologi, dimulai pengumpulan data saham dan berita, praproses data, ekstraksi fitur indikator teknikal dan sentimen serta analisis data, selanjutnya pengembangan model. Serangkaian uji coba parameter crossover dan mutasi GA memberi hasil optimum pencarian kombinatorik hyper-parameter model dengan accuracy 81.63% dan f1-score 82.21%. Evaluasi model terhadap kombinasi jenis dataset mampu meningkatkan accuracy prediksi dari 75.91% menajdi 81.63%, dan f1-score dari 77.56% menjadi 82.21%. Terhadap evaluasi trading, metode yang diusulkan terbukti memberi return yang fantastis sebesar 121.27% dalam setahun, dengan nilai maximum drawdown yang paling kecil juga nilai sharpe ratio yang tinggi. Evaluasi tersebut melampaui hasil penelitian serupa terdahulu, bahkan jauh diatas performa pergerakan saham itu sendiri terindikasi melalui strategi buy & hold

A refined measure of conditional maximum drawdown

Risks associated to maximum drawdown have been recently formalized as the tail mean of the maximum drawdown distribution, called Conditional Expected Drawdown (CED). In fact, the special case of average maximum drawdown is widely used in the fund management industry also in association to performance management. It lacks relevant information on worst case scenarios over a fixed horizon. Formulating a refined version of CED, we are able to add this piece of information to the risk measurement of drawdown, and then get a risk measure for processes that preserves all the good properties of CED but following more prudential regulatory and management assessments, also in term of marginal risk contribution attributed to factors. As a special application, we consider the conditioning information given by the all time minimum of cumulative returns.

Multi-time state mean-variance model in continuous time

The objective of the continuous time mean-variance model is to minimize the variance (risk) of an investment portfolio with a given mean at the terminal time. However, the investor can stop the investment plan at any time before the terminal time. To solve this problem, we consider to minimize the variances of the investment portfolio in the multi-time state. The advantage of this multi-time state mean-variance model is the minimization of the risk of the investment portfolio within the investment period. To obtain the optimal strategy of the model, we introduce a sequence of Riccati equations, which are connected by jump boundary conditions. In addition, we establish the relationships between the means and variances in the multi-time state mean-variance model. Furthermore, we use an example to verify that the variances of the multi-time state can affect the average of Maximum-Drawdown of the investment portfolio.