Data Analysis & Utility Calculators

Available Tools

• Unit Converter: Convert between various measurement units
• QR Code Generator: Generate QR codes from text or URLs
• Assignment Problem Solver: Solve assignment problems using the Hungarian algorithm
• Statistics Calculator: Calculate basic descriptive statistics from a dataset
• Regression Calculator: Perform linear regression analysis on x,y data
• NPV Calculator: Calculate net present value and internal rate of return
• Correlation Heatmap: Compute and visualize Pearson correlation matrix
• Multiple Regression Analysis: Fit multiple linear regression model with diagnostics
• PCA Dimension Reduction: Reduce dimensions using Principal Component Analysis
• Feature Importance: Rank predictors by importance to the target variable
• K-Means Clustering: Partition data into K clusters using K-Means algorithm
• Hierarchical Clustering: Agglomerative clustering with linkage methods
• DBSCAN Clustering: Density-based clustering with noise detection
• Interpolation Calculator: Perform linear interpolation or extrapolation from data points
• Data Profiler: Compute descriptive statistics for numeric datasets
• ROI Calculator: Calculate return on investment and payback period
• LCC Calculator: Calculate life cycle cost for assets and equipment
• Depreciation Calculator: Calculate asset depreciation using straight-line or declining-balance method
• Weighted Score Calculator: Evaluate alternatives using weighted multi-criteria scoring
• Moving Average (SMA/EMA) Calculator: Smooth a data series and reveal trends using a Simple or Exponential Moving Average

Quantitative Utilities for Engineers, Analysts, and Students

This domain collects lightweight calculators for the recurring quantitative chores that fall outside any single engineering discipline. Engineers use them to check a correlation between a process variable and yield, fit a linear or multiple regression, or compute net present value for a capital project before escalating to a full financial model. Data analysts reach for k-means, DBSCAN, hierarchical clustering, PCA, and feature importance as exploratory passes before committing to a modelling pipeline. Cost engineers combine depreciation, ROI, NPV, and life-cycle cost routines to size the arithmetic behind a proposal. Students learning statistics or machine-learning fundamentals use the descriptive statistics, data profiler, interpolation, and weighted-score tools to verify textbook exercises. Each calculator runs entirely in the browser on pasted numeric input, returning the same canonical quantities a spreadsheet or scripting session would produce, without configuration or a server round trip.

Intended Users and Boundaries

The target user is the practitioner who needs a defensible number in minutes: the data analyst triaging a CSV before pulling it into a notebook, the reliability engineer running a quick correlation or regression on warranty data, the cost engineer sketching an NPV or ROI for a capital request, and the student cross-checking a k-means or PCA homework result. The calculators serve this exploratory, single-file tier. They are not a substitute for a full data-science workflow with versioned features, cross-validated models, and reproducible pipelines; they do not host production ML serving, drift monitoring, or batch inference; and they must not be used for regulated financial valuation, audit-grade discounted-cash-flow models, or statutory tax filings. When a decision carries audit, safety, or material financial weight, these outputs belong in a working paper that feeds a validated tool, not as the final authority.

Methods and References

Descriptive statistics follow the standard sample-moment definitions; the F distribution used for regression significance traces to Fisher and Snedecor, while correlation uses the Pearson product-moment coefficient (Pearson, 1895). Linear and multiple regression are fitted by ordinary least squares with the usual normal-equation solution. K-means iterates the Lloyd (1957/1982) assignment-update algorithm; DBSCAN implements the density-reachability formulation of Ester, Kriegel, Sander, and Xu (1996); hierarchical clustering uses agglomerative linkage on a pairwise distance matrix. Principal component analysis follows Pearson (1901) and Hotelling (1933) via covariance-matrix eigendecomposition. Net present value uses the textbook discounted-cash-flow identity, and life-cycle costing aligns with ISO 15686-5. Unit conversions reference the SI base-unit definitions maintained by the BIPM. QR code generation conforms to ISO/IEC 18004, and the assignment tool implements the Kuhn-Munkres (Hungarian) algorithm.

Out of Scope

These tools deliberately stop at the exploratory and teaching tier. They do not ingest streaming data, schedule jobs, or host production data pipelines, and they do not train, tune, or serve machine-learning models at scale; artefacts such as cross-validation folds, hyperparameter search, model registries, and inference endpoints belong in a dedicated platform. Time-series forecasting, ARIMA or state-space models, survival analysis, Bayesian inference, and advanced econometrics (panel data, instrumental variables, cointegration) are outside the scope of the regression and statistics calculators. The depreciation tool returns textbook straight-line, declining-balance, and sum-of-years-digits figures and is not a substitute for jurisdiction-specific tax treatment, MACRS schedules, or IFRS/GAAP-compliant accounting entries. NPV and ROI outputs are decision-support sketches, not audit-grade valuation reports, and should be reproduced in a validated financial model before any binding commitment.