Closes #020 - Ticker Data Real-Time Alignment & ML Handbook Integration

components/modules/crypto/CryptoMathModal.tsx

@@ -26,8 +26,8 @@ export default function CryptoMathModal({ isOpen, onClose }: CryptoMathModalProp
   if (!isOpen) return null;
 
   return (
-    <div className="fixed inset-0 z-50 flex items-center justify-center bg-slate-950/85 backdrop-blur-md p-4 sm:p-6 md:p-8 animate-fade-in">
-      <div className="bg-slate-900 border border-slate-800/80 rounded-3xl w-full max-w-4xl h-[80vh] flex flex-col overflow-hidden shadow-2xl relative text-slate-300">
+    <div className="fixed inset-0 z-50 overflow-y-auto bg-slate-950/85 backdrop-blur-md flex items-start justify-center p-4 sm:p-6 md:p-8 animate-fade-in">
+      <div className="bg-slate-900 border border-slate-800/80 rounded-3xl w-full max-w-4xl my-auto max-h-full flex flex-col overflow-hidden shadow-2xl relative text-slate-300">
         
         {/* Modal Header */}
         <div className="flex justify-between items-center px-6 py-4 bg-slate-950/45 border-b border-slate-800/60">
@@ -181,6 +181,70 @@ export default function CryptoMathModal({ isOpen, onClose }: CryptoMathModalProp
             </div>
           </div>
 
+          {/* Section G: Ensemble Estimator Specifications */}
+          <div className="space-y-3">
+            <h4 className="text-xs font-bold text-cyan-400 uppercase tracking-wider font-mono">G. Ensemble Estimator Specifications</h4>
+            <p className="text-xs leading-relaxed text-slate-400">
+              The Walk-Forward Ensemble Radar aggregates forecasting signals from 5 independent machine learning estimators optimized for distinct predictive roles across three temporal horizons:
+            </p>
+            <div className="bg-slate-950/40 p-5 rounded-2xl border border-slate-800/60 space-y-4">
+              <div className="space-y-1">
+                <span className="text-xs font-semibold text-slate-200 block">1. Random Forest (RF)</span>
+                <p className="text-xs text-slate-400 leading-relaxed">
+                  Utilizes bootstrap aggregation (bagging) of uncorrelated decision trees to map non-linear feature interactions. It constructs a robust ensemble prediction:
+                </p>
+                <BlockMath math="\hat{P}_{\text{RF}}(y=1 \mid \mathbf{x}) = \frac{1}{B} \sum_{b=1}^{B} f_b(\mathbf{x})" />
+                <p className="text-[10px] text-slate-500 font-mono">
+                  Optimized for multi-regime boundary separation and filtering out high-volatility futures noise.
+                </p>
+              </div>
+
+              <div className="space-y-1">
+                <span className="text-xs font-semibold text-slate-200 block">2. XGBoost / Gradient Boosting (GB)</span>
+                <p className="text-xs text-slate-400 leading-relaxed">
+                  Fits sequential decision trees to minimize the residual classification loss via gradient descent:
+                </p>
+                <BlockMath math="\mathcal{L}^{(t)} = \sum_{i=1}^{n} l\left(y_i, \hat{y}_i^{(t-1)} + f_t(\mathbf{x}_i)\right) + \Omega(f_t)" />
+                <p className="text-[10px] text-slate-500 font-mono">
+                  Highly responsive to short-term micro-trends, making it the primary signal anchor for the T+1 horizon.
+                </p>
+              </div>
+
+              <div className="space-y-1">
+                <span className="text-xs font-semibold text-slate-200 block">3. Logistic Regression with ElasticNet (LR)</span>
+                <p className="text-xs text-slate-400 leading-relaxed">
+                  Serves as the linear baseline anchor, regularized with combined L1 (Lasso) and L2 (Ridge) penalties:
+                </p>
+                <BlockMath math="\min_{\mathbf{w}, c} \frac{1}{n} \sum_{i=1}^{n} \log\left(1 + e^{-y_i (\mathbf{w}^T \mathbf{x}_i + c)}\right) + r \lambda \|\mathbf{w}\|_1 + \frac{1-r}{2} \lambda \|\mathbf{w}\|_2^2" />
+                <p className="text-[10px] text-slate-500 font-mono">
+                  Prevents wild regime-extrapolation decay and ensures structural stability during major trend shifts.
+                </p>
+              </div>
+
+              <div className="space-y-1">
+                <span className="text-xs font-semibold text-slate-200 block">4. Support Vector Machine (SVM)</span>
+                <p className="text-xs text-slate-400 leading-relaxed">
+                  Projects the feature space into a high-dimensional Hilbert space using a Radial Basis Function (RBF) kernel:
+                </p>
+                <BlockMath math="K(\mathbf{x}_i, \mathbf{x}_j) = \exp\left(-\gamma \|\mathbf{x}_i - \mathbf{x}_j\|^2\right)" />
+                <p className="text-[10px] text-slate-500 font-mono">
+                  Isolates non-linear hyperplane separation boundaries, targeting multi-dimensional trend-reversal thresholds for the T+5 horizon.
+                </p>
+              </div>
+
+              <div className="space-y-1">
+                <span className="text-xs font-semibold text-slate-200 block">5. Multi-Layer Perceptron (MLP)</span>
+                <p className="text-xs text-slate-400 leading-relaxed">
+                  A deep feedforward neural network mapping complex cross-correlations across hidden layers using backpropagation:
+                </p>
+                <BlockMath math="\mathbf{a}^{(l)} = \sigma\left(\mathbf{W}^{(l)} \mathbf{a}^{(l-1)} + \mathbf{b}^{(l)}\right)" />
+                <p className="text-[10px] text-slate-500 font-mono">
+                  Extracts intricate temporal patterns and deep feature interactions, optimized for the medium-term T+10 forecasting horizon.
+                </p>
+              </div>
+            </div>
+          </div>
+
         </div>
       </div>
     </div>