JPS SRC

Dalawang klase ng sistema ngayon:
1. Sistema na bubugbugin ka hanggang maging mahusay ka.
2. Sistema na bubugbugin ka lang.

📊 STAT TRIVIA | Panu naman kapag maraming pattern sa isang graph

Ano ang napapansin mo sa graph na ito? 👀

May tatlong panels at sa bawat panel, may tatlong magkakaibang lines.

Sa unang tingin, parang simpleng line graph lang. Pero sa statistical analysis, may mas interesting na tanong:

Magkakapareho ba ang pattern ng pagbabago ng bawat group habang lumilipas ang panahon?

🔎 Tingnan natin ang pattern

Sa Panel A, makikita na generally mataas ang values ng tatlong trajectories. May initial increase, may ilang fluctuations, at pagkatapos ay nagkakaroon ng relative stabilization.

Sa Panel B, mas magkakaiba ang behavior ng mga lines. May isang trajectory na mabilis tumaas sa early period, habang ang iba ay nasa mas mababang level at nagbabago naman sa later period.

Sa Panel C, mas kapansin-pansin ang difference. May trajectory na nananatiling relatively low, habang ang ibang lines ay nagpapakita ng mas malaking increase toward the later observation days.

📐 Saan pumapasok ang statistics?

Dito nagiging useful ang time-series analysis at mathematical modeling.

Sa research na ito, ginamit ang RStudio para:

i-organize at i-visualize ang reconstructed data;
tingnan ang temporal patterns;
ikumpara ang iba't ibang trajectories;
at mag-fit ng mathematical models sa mga observed patterns.

Hindi lang natin tinitingnan kung alin ang pinakamataas.

Mas mahalaga ring tingnan:

📈 Gaano kabilis ang pagbabago?
📉 May slowing down ba?
🔄 May pagbabago ba sa pattern over time?
📊 Pareho ba ang mathematical behavior ng bawat trajectory?

💡 Simple example

Imagine may tatlong estudyante na nagre-review for an exam.

Student A: mabilis ang improvement sa simula, pagkatapos halos stable na.

Student B: mabagal sa simula pero mas mabilis ang improvement sa later days.

Student C: mababa ang improvement pero relatively consistent.

Pare-pareho silang “nag-i-improve,” pero magkaiba ang pattern ng improvement.

Ganito rin ang idea ng mathematical modeling:

Hindi lang natin pinagkukumpara ang values; sinusubukan nating maintindihan ang mathematical pattern ng pagbabago.

At gamit ang RStudio, maaaring i-fit at ikumpara ang mathematical models upang makita kung alin ang mas angkop na kumatawan sa bawat trajectory.

📌 Research note:
Ang figure na ito ay bahagi ng isang ongoing mathematical modeling study. Ang specific research variables at deeper experimental interpretation ay hindi muna dini-disclose.

Question for you:
👀 Kung ikaw ang titingin sa graph, alin sa tatlong panels ang may pinaka-interesting na pagbabago sa trajectory—A, B, or C?

📊 STAT TRIVIA | Paano Binabasa ng Math ang Isang Graph?

Akala mo simpleng line graph lang? 👀

Pero sa research, bawat point sa graph ay maaaring maging bahagi ng isang mathematical model.

Tingnan ang figure. Mapapansin natin na ang trajectory ay hindi simpleng straight line:

📈 mabilis na increase
➡️ halos stable na bahagi
➡️ panibagong increase
➡️ paglapit sa isang mas mataas na level

🤔 So anong statistics/math ang ginagamit dito?

Sa research na ito, gumamit kami ng RStudio para i-process at i-analyze ang reconstructed time-series data.

Hindi lang ordinaryong descriptive statistics ang ginawa. Gumamit din ng mathematical model fitting upang ikumpara kung aling mathematical pattern ang mas angkop sa trajectory.

Kabilang dito ang:

🔹 Logistic Model – ginagamit para i-represent ang growth pattern na maaaring magkaroon ng slowing or leveling-off behavior.

🔹 Gompertz Model – isa pang nonlinear model na maaaring mag-represent ng asymmetric growth pattern.

🔹 Piecewise Model – useful kapag may bahagi ng trajectory na maaaring magkaroon ng magkaibang pattern o slope bago at pagkatapos ng isang transition point.

🧠 Simple example

Imagine may data tayo ng isang halaman:

Day 1 → 5 cm
Day 2 → 8 cm
Day 3 → 9 cm
Day 4 → 9 cm
Day 5 → 15 cm

Hindi lang natin sasabihin:

“Tumaas ang halaman.”

Gamit ang mathematical modeling, puwede nating itanong:

“Anong mathematical model ang pinaka-angkop para i-describe ang pattern ng pagbabago?”

At dito pumapasok ang RStudio—ginagamit ito para i-fit ang mathematical models sa data at ikumpara ang kanilang performance.

Para sa model comparison, gumamit din ng measures tulad ng RMSE, MAE, AIC, at BIC. Sa simple terms, ginagamit ang mga ito para makatulong na makita kung gaano kalapit ang model sa observed/reconstructed data at kung paano nagkukumpara ang iba't ibang mathematical formulations.

📌 Important: Ang figure na ito ay bahagi ng isang ongoing mathematical modeling research.

Kaya next time na makakita ka ng graph, huwag lang tanungin kung “tumataas ba?” 😉
Tanungin mo rin: “Anong mathematical pattern ang nasa likod nito?”

Paano kaya natin mababawasan ang maraming variables nang hindi nawawala ang mahahalagang impormasyon? 🤔

Dito pumapasok ang Principal Component Analysis (PCA)—isang statistical technique na ginagamit para i-transform ang maraming variables into a smaller set of principal components, habang pinapanatili hangga’t maaari ang variation/information na nasa data.

💡 Quick Trivia:
Ang PC1 (Principal Component 1) ang kumukuha ng pinakamalaking variance sa dataset, habang ang PC2 ang susunod na pinakamalaking variance at orthogonal sa PC1.
📌 Common applications:
🔹 Data visualization
🔹 Dimensionality reduction
🔹 Feature extraction
🔹 Exploratory data analysis
🔹 Machine learning preprocessing

Sa madaling salita:
Maraming variables → PCA → Mas kaunting dimensions → Mahahalagang patterns retained. 📈

Statistics is not just about numbers—it’s about finding the structure hidden within the data. 🔍📊

Looking for a Reliable Statistical & Research Partner? Let’s Work Together. 🤝

DR.JPS Statistical and Research Consultancy (JPSSRC) is a Philippine-based, registered research consultancy that is open to partnering with organizations locally and internationally for statistical, research, and data-driven projects.

We support organizations that need reliable external expertise in:
📊 Statistical Analysis & Interpretation
🔬 Quantitative Research
📈 Data Analysis & Reporting
🧮 Statistical Modeling
📝 Research Design & Methodology
💻 Research Data & Analytical Support
🎓 Educational & Institutional Research
👥 Statistics & Research Training

We are open to project-based, remote, contractual, outsourced, and longer-term engagements, depending on your organization’s needs.

Why work with JPSSRC?
We aim to provide practical, evidence-based, and professionally documented statistical and research support that can help organizations make better decisions, strengthen research outputs, and turn data into meaningful insights.
We are particularly interested in connecting with:
🌐 International companies and organizations
🏢 Research and consulting firms
🎓 Universities and research institutions
💻 Technology and data-driven companies
🌱 NGOs and development organizations
📊 Organizations with ongoing or upcoming research and data projects

If your organization needs an external statistical or research partner, JPSSRC would be glad to explore how we can support your team.
📩 [email protected]
📞 +63 977 883 3026

🌏 Open to local and international collaborations.
Statistics • Research • Data Analytics • Quantitative Solutions

This time, we worked on the Test of Normality as part of the preliminary assessment before proceeding with the inferential statistical analyses.

💡 Quick Stat Trivia:
Did you know that before applying statistical procedures such as regression, researchers need to check whether the assumptions of the intended analysis are reasonably satisfied?
For regression analysis, one important diagnostic is the normality of the regression residuals. This can be assessed using both:
✔️ Shapiro–Wilk Test – provides a formal statistical test of normality
✔️ Q–Q Plot – provides a visual assessment of how closely the residuals follow a normal distribution
📈 In actual research, statistical tests should not be interpreted in isolation. With sufficiently large samples, even small departures from normality may become statistically significant. Thus, graphical diagnostics and the overall pattern of the residuals should also be considered when evaluating the assumptions of a model.

At JPSSRC, we don't just generate statistical outputs—we help ensure that the appropriate statistical procedures, diagnostics, and interpretations are considered throughout the research process.

🔒 Research details and client-specific findings are intentionally withheld to protect data confidentiality.

Another research milestone completed. Another PhD journey one step closer. 🎓📊

We recently completed the statistical computation and analysis for one of our PhD clients, using RStudio.

The graph shown is an example of an Observed vs. Predicted Values visualization from a regression analysis.

💡 Why is this graph useful?
It provides a visual way to examine how well the statistical model's predicted values correspond with the actual observations. It is one of the useful graphical outputs researchers can use alongside the numerical regression results when evaluating a predictive model.

For this study, the regression analysis showed that the predictors made a statistically significant contribution to the outcome being studied.

🔐 Research variables, datasets, and detailed numerical results are withheld to maintain client confidentiality.

Statistical computation completed. Researcher ready for the next stage. ✔️

RStudio • Statistical Analysis • Research Data Analysis • Thesis & Dissertation Support

📩 Need help turning your research data into meaningful statistical results? Send us a message.

✨ READY TO SERVE!

DR.JPS Statistical and Research Consultancy (JPSSRC) is ready to serve you with professional, reliable, and quality consultancy services.

📩 Let’s work together!
Email us at [email protected]
JPSSRC — Knowledge Sharing Platform.

📚 Every child deserves the opportunity to succeed in Mathematics.

Sa JPSSRC, naniniwala kami na ang pananaliksik ay hindi dapat manatili lamang sa papel—dapat itong maging konkretong solusyon sa mga hamon ng edukasyon.
Bilang bahagi ng aming Education Advocacy, layunin naming isagawa ang Phase 1 Pilot Implementation ng SLIM (Spatial Learning Instructional Model) Framework—isang research-based instructional framework na idinisenyo upang matulungan ang mga mag-aaral na nahihirapan sa Mathematics—sa isang barangay bilang unang hakbang tungo sa mas malawak na implementasyon.

🤝 Inaanyayahan namin ang mga indibidwal, organisasyon, kumpanya, foundations, at institusyon na makiisa sa aming advocacy. Sa pamamagitan ng partnership, sponsorship, volunteerism, o anumang anyo ng suporta, sama-sama nating mabibigyan ng mas magandang oportunidad sa pagkatuto ang ating mga kabataan.

Together, let's transform research into real community impact.

📞 0977-883-3026
📧 [email protected]

JPSSRC – Knowledge Sharing Platform