[{"data":1,"prerenderedAt":918},["ShallowReactive",2],{"content-query-YTSCRCj0jS":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"heading":10,"prompt":11,"tags":15,"files":17,"nav":6,"presets":18,"gallery":35,"body":37,"_type":911,"_id":912,"_source":913,"_file":914,"_stem":915,"_extension":916,"sitemap":917},"/tools/trendline-calculator","tools",false,"","Trendline Calculator for Excel & CSV","Add and compare trendlines online from Excel and CSV data. Fit linear, polynomial, exponential, and other trends with AI.","Trendline Calculator",{"prefix":12,"label":13,"placeholder":14},"Fit a trendline","Describe the trendline analysis you want to run","e.g. fit linear and exponential trendlines to monthly sales data, compare R², project 6 months forward",[16],"statistics",true,[19,25,30],{"label":20,"prompt":21,"dataset_url":22,"dataset_title":23,"dataset_citation":24},"GDP per capita trend","fit linear and exponential trendlines to GDP per capita over time for the United States; compare R²; plot both fits with 95% confidence bands; report annual growth rate from exponential fit; project 10 years forward","https://ourworldindata.org/grapher/gdp-per-capita-worldbank.csv","GDP per capita","Our World in Data",{"label":26,"prompt":27,"dataset_url":28,"dataset_title":29,"dataset_citation":24},"Internet users growth","fit exponential and polynomial (degree 2) trendlines to the share of population using the internet over time for the world; compare R²; annotate inflection point; project to 2030","https://ourworldindata.org/grapher/share-of-individuals-using-the-internet.csv","Share of population using the internet",{"label":31,"prompt":32,"dataset_url":33,"dataset_title":34,"dataset_citation":24},"Solar energy growth","fit exponential trendline to solar energy consumption over time; extract doubling time from exponential growth rate; plot fit with 95% confidence band; annotate projected capacity for next 5 years","https://ourworldindata.org/grapher/solar-energy-consumption.csv","Solar energy consumption",[36],"/img/tools/trendline-calculator.png",{"type":38,"children":39,"toc":900},"root",[40,49,99,118,124,203,209,362,367,373,522,528,651,657,719,725,760,766,823,847,857,883],{"type":41,"tag":42,"props":43,"children":45},"element","h2",{"id":44},"what-is-a-trendline",[46],{"type":47,"value":48},"text","What Is a Trendline?",{"type":41,"tag":50,"props":51,"children":52},"p",{},[53,55,61,63,68,70,75,77,83,85,90,92,97],{"type":47,"value":54},"A ",{"type":41,"tag":56,"props":57,"children":58},"strong",{},[59],{"type":47,"value":60},"trendline",{"type":47,"value":62}," is a line or curve fitted to a dataset to summarize the overall direction and rate of change. It reduces noisy, observation-by-observation data to a smooth mathematical function that captures the underlying pattern — and allows you to project that pattern forward into the future. The choice of trendline type encodes a hypothesis about how the variable changes: a ",{"type":41,"tag":56,"props":64,"children":65},{},[66],{"type":47,"value":67},"linear trendline",{"type":47,"value":69}," assumes a constant rate of change (y = a + bx); an ",{"type":41,"tag":56,"props":71,"children":72},{},[73],{"type":47,"value":74},"exponential trendline",{"type":47,"value":76}," assumes a constant ",{"type":41,"tag":78,"props":79,"children":80},"em",{},[81],{"type":47,"value":82},"percentage",{"type":47,"value":84}," rate of change (y = ae^(bx)), appropriate for populations, compound growth, and viral spread; a ",{"type":41,"tag":56,"props":86,"children":87},{},[88],{"type":47,"value":89},"polynomial trendline",{"type":47,"value":91}," captures curves with one or more bends (quadratic, cubic); and a ",{"type":41,"tag":56,"props":93,"children":94},{},[95],{"type":47,"value":96},"moving average",{"type":47,"value":98}," smooths out short-term fluctuations without committing to a parametric model.",{"type":41,"tag":50,"props":100,"children":101},{},[102,104,109,111,116],{"type":47,"value":103},"The primary goodness-of-fit metric is ",{"type":41,"tag":56,"props":105,"children":106},{},[107],{"type":47,"value":108},"R²",{"type":47,"value":110}," (coefficient of determination): the proportion of variance in the data explained by the trendline, ranging from 0 (no fit) to 1 (perfect fit). A higher-degree polynomial will always achieve higher R² than a linear fit on the same data — but that doesn't mean it's the better model. A quadratic may capture a real curve or may simply be overfitting random noise. The correct approach is to choose the trendline type based on ",{"type":41,"tag":56,"props":112,"children":113},{},[114],{"type":47,"value":115},"subject-matter knowledge",{"type":47,"value":117}," (does exponential growth make physical sense here?) and then use R² to quantify how well that model fits, rather than choosing the type that maximizes R².",{"type":41,"tag":42,"props":119,"children":121},{"id":120},"how-it-works",[122],{"type":47,"value":123},"How It Works",{"type":41,"tag":125,"props":126,"children":127},"ol",{},[128,153,168],{"type":41,"tag":129,"props":130,"children":131},"li",{},[132,137,139,144,146,151],{"type":41,"tag":56,"props":133,"children":134},{},[135],{"type":47,"value":136},"Upload your data",{"type":47,"value":138}," — provide a CSV or Excel file with an ",{"type":41,"tag":56,"props":140,"children":141},{},[142],{"type":47,"value":143},"x",{"type":47,"value":145}," column (time, concentration, distance, or any numeric variable) and a ",{"type":41,"tag":56,"props":147,"children":148},{},[149],{"type":47,"value":150},"y",{"type":47,"value":152}," column (the measured outcome). One row per observation.",{"type":41,"tag":129,"props":154,"children":155},{},[156,161,163],{"type":41,"tag":56,"props":157,"children":158},{},[159],{"type":47,"value":160},"Describe the analysis",{"type":47,"value":162}," — e.g. ",{"type":41,"tag":78,"props":164,"children":165},{},[166],{"type":47,"value":167},"\"fit linear and exponential trendlines to the 'month' and 'revenue' columns; compare R²; project 6 months forward; plot with confidence bands\"",{"type":41,"tag":129,"props":169,"children":170},{},[171,176,178,185,187,193,195,201],{"type":41,"tag":56,"props":172,"children":173},{},[174],{"type":47,"value":175},"Get full results",{"type":47,"value":177}," — the AI writes Python code using ",{"type":41,"tag":179,"props":180,"children":182},"a",{"href":181},"https://docs.scipy.org/doc/scipy/reference/optimize.html",[183],{"type":47,"value":184},"scipy.optimize",{"type":47,"value":186},", ",{"type":41,"tag":179,"props":188,"children":190},{"href":189},"https://numpy.org/doc/stable/reference/generated/numpy.polyfit.html",[191],{"type":47,"value":192},"numpy.polyfit",{"type":47,"value":194},", and ",{"type":41,"tag":179,"props":196,"children":198},{"href":197},"https://plotly.com/python/",[199],{"type":47,"value":200},"Plotly",{"type":47,"value":202}," to overlay multiple trendlines, display R² in the legend, shade the projection region, and report the fitted equation",{"type":41,"tag":42,"props":204,"children":206},{"id":205},"required-data-format",[207],{"type":47,"value":208},"Required Data Format",{"type":41,"tag":210,"props":211,"children":212},"table",{},[213,237],{"type":41,"tag":214,"props":215,"children":216},"thead",{},[217],{"type":41,"tag":218,"props":219,"children":220},"tr",{},[221,227,232],{"type":41,"tag":222,"props":223,"children":224},"th",{},[225],{"type":47,"value":226},"Column",{"type":41,"tag":222,"props":228,"children":229},{},[230],{"type":47,"value":231},"Description",{"type":41,"tag":222,"props":233,"children":234},{},[235],{"type":47,"value":236},"Example",{"type":41,"tag":238,"props":239,"children":240},"tbody",{},[241,290,329],{"type":41,"tag":218,"props":242,"children":243},{},[244,254,259],{"type":41,"tag":245,"props":246,"children":247},"td",{},[248],{"type":41,"tag":249,"props":250,"children":252},"code",{"className":251},[],[253],{"type":47,"value":143},{"type":41,"tag":245,"props":255,"children":256},{},[257],{"type":47,"value":258},"Independent variable — time, dose, distance",{"type":41,"tag":245,"props":260,"children":261},{},[262,268,269,275,276,282,284],{"type":41,"tag":249,"props":263,"children":265},{"className":264},[],[266],{"type":47,"value":267},"1",{"type":47,"value":186},{"type":41,"tag":249,"props":270,"children":272},{"className":271},[],[273],{"type":47,"value":274},"2",{"type":47,"value":186},{"type":41,"tag":249,"props":277,"children":279},{"className":278},[],[280],{"type":47,"value":281},"3",{"type":47,"value":283}," … (month) or ",{"type":41,"tag":249,"props":285,"children":287},{"className":286},[],[288],{"type":47,"value":289},"2020-01-01",{"type":41,"tag":218,"props":291,"children":292},{},[293,301,306],{"type":41,"tag":245,"props":294,"children":295},{},[296],{"type":41,"tag":249,"props":297,"children":299},{"className":298},[],[300],{"type":47,"value":150},{"type":41,"tag":245,"props":302,"children":303},{},[304],{"type":47,"value":305},"Dependent variable — count, sales, concentration",{"type":41,"tag":245,"props":307,"children":308},{},[309,315,316,322,323],{"type":41,"tag":249,"props":310,"children":312},{"className":311},[],[313],{"type":47,"value":314},"12.3",{"type":47,"value":186},{"type":41,"tag":249,"props":317,"children":319},{"className":318},[],[320],{"type":47,"value":321},"18.7",{"type":47,"value":186},{"type":41,"tag":249,"props":324,"children":326},{"className":325},[],[327],{"type":47,"value":328},"31.2",{"type":41,"tag":218,"props":330,"children":331},{},[332,341,346],{"type":41,"tag":245,"props":333,"children":334},{},[335],{"type":41,"tag":249,"props":336,"children":338},{"className":337},[],[339],{"type":47,"value":340},"group",{"type":41,"tag":245,"props":342,"children":343},{},[344],{"type":47,"value":345},"Optional: series label for multi-series comparison",{"type":41,"tag":245,"props":347,"children":348},{},[349,355,356],{"type":41,"tag":249,"props":350,"children":352},{"className":351},[],[353],{"type":47,"value":354},"Product A",{"type":47,"value":186},{"type":41,"tag":249,"props":357,"children":359},{"className":358},[],[360],{"type":47,"value":361},"Product B",{"type":41,"tag":50,"props":363,"children":364},{},[365],{"type":47,"value":366},"Any column names work — describe them in your prompt. Dates are automatically converted to numeric time indices for fitting.",{"type":41,"tag":42,"props":368,"children":370},{"id":369},"interpreting-the-results",[371],{"type":47,"value":372},"Interpreting the Results",{"type":41,"tag":210,"props":374,"children":375},{},[376,392],{"type":41,"tag":214,"props":377,"children":378},{},[379],{"type":41,"tag":218,"props":380,"children":381},{},[382,387],{"type":41,"tag":222,"props":383,"children":384},{},[385],{"type":47,"value":386},"Output",{"type":41,"tag":222,"props":388,"children":389},{},[390],{"type":47,"value":391},"What it means",{"type":41,"tag":238,"props":393,"children":394},{},[395,410,426,442,458,474,490,506],{"type":41,"tag":218,"props":396,"children":397},{},[398,405],{"type":41,"tag":245,"props":399,"children":400},{},[401],{"type":41,"tag":56,"props":402,"children":403},{},[404],{"type":47,"value":108},{"type":41,"tag":245,"props":406,"children":407},{},[408],{"type":47,"value":409},"Proportion of variance explained — closer to 1 = better fit for that trendline type",{"type":41,"tag":218,"props":411,"children":412},{},[413,421],{"type":41,"tag":245,"props":414,"children":415},{},[416],{"type":41,"tag":56,"props":417,"children":418},{},[419],{"type":47,"value":420},"Slope (linear)",{"type":41,"tag":245,"props":422,"children":423},{},[424],{"type":47,"value":425},"Rate of change per unit x — e.g. $4,200 revenue per month",{"type":41,"tag":218,"props":427,"children":428},{},[429,437],{"type":41,"tag":245,"props":430,"children":431},{},[432],{"type":41,"tag":56,"props":433,"children":434},{},[435],{"type":47,"value":436},"Growth rate (exponential)",{"type":41,"tag":245,"props":438,"children":439},{},[440],{"type":47,"value":441},"b in y = ae^(bx) — multiply by 100 for % growth per unit x",{"type":41,"tag":218,"props":443,"children":444},{},[445,453],{"type":41,"tag":245,"props":446,"children":447},{},[448],{"type":41,"tag":56,"props":449,"children":450},{},[451],{"type":47,"value":452},"Doubling time",{"type":41,"tag":245,"props":454,"children":455},{},[456],{"type":47,"value":457},"ln(2) / b — how long until the value doubles at exponential rate b",{"type":41,"tag":218,"props":459,"children":460},{},[461,469],{"type":41,"tag":245,"props":462,"children":463},{},[464],{"type":41,"tag":56,"props":465,"children":466},{},[467],{"type":47,"value":468},"Polynomial coefficients",{"type":41,"tag":245,"props":470,"children":471},{},[472],{"type":47,"value":473},"β₀, β₁, β₂ … — interpret the quadratic term for curvature direction",{"type":41,"tag":218,"props":475,"children":476},{},[477,485],{"type":41,"tag":245,"props":478,"children":479},{},[480],{"type":41,"tag":56,"props":481,"children":482},{},[483],{"type":47,"value":484},"Moving average",{"type":41,"tag":245,"props":486,"children":487},{},[488],{"type":47,"value":489},"Smoothed value at each point — no equation, useful for visualizing cycles",{"type":41,"tag":218,"props":491,"children":492},{},[493,501],{"type":41,"tag":245,"props":494,"children":495},{},[496],{"type":41,"tag":56,"props":497,"children":498},{},[499],{"type":47,"value":500},"Confidence band",{"type":41,"tag":245,"props":502,"children":503},{},[504],{"type":47,"value":505},"Range of plausible mean y at each x — widens as you extrapolate further",{"type":41,"tag":218,"props":507,"children":508},{},[509,517],{"type":41,"tag":245,"props":510,"children":511},{},[512],{"type":41,"tag":56,"props":513,"children":514},{},[515],{"type":47,"value":516},"Projection",{"type":41,"tag":245,"props":518,"children":519},{},[520],{"type":47,"value":521},"Extrapolated trendline beyond the data — always treat as a best-case linear estimate",{"type":41,"tag":42,"props":523,"children":525},{"id":524},"example-prompts",[526],{"type":47,"value":527},"Example Prompts",{"type":41,"tag":210,"props":529,"children":530},{},[531,547],{"type":41,"tag":214,"props":532,"children":533},{},[534],{"type":41,"tag":218,"props":535,"children":536},{},[537,542],{"type":41,"tag":222,"props":538,"children":539},{},[540],{"type":47,"value":541},"Scenario",{"type":41,"tag":222,"props":543,"children":544},{},[545],{"type":47,"value":546},"What to type",{"type":41,"tag":238,"props":548,"children":549},{},[550,567,584,601,617,634],{"type":41,"tag":218,"props":551,"children":552},{},[553,558],{"type":41,"tag":245,"props":554,"children":555},{},[556],{"type":47,"value":557},"Compare all types",{"type":41,"tag":245,"props":559,"children":560},{},[561],{"type":41,"tag":249,"props":562,"children":564},{"className":563},[],[565],{"type":47,"value":566},"fit linear, exponential, and polynomial (degree 2) trendlines; compare R² for each; plot overlaid on scatter",{"type":41,"tag":218,"props":568,"children":569},{},[570,575],{"type":41,"tag":245,"props":571,"children":572},{},[573],{"type":47,"value":574},"Forecast forward",{"type":41,"tag":245,"props":576,"children":577},{},[578],{"type":41,"tag":249,"props":579,"children":581},{"className":580},[],[582],{"type":47,"value":583},"fit exponential trendline to monthly sales; project 12 months forward; shade projection region; report projected values",{"type":41,"tag":218,"props":585,"children":586},{},[587,592],{"type":41,"tag":245,"props":588,"children":589},{},[590],{"type":47,"value":591},"Growth rate",{"type":41,"tag":245,"props":593,"children":594},{},[595],{"type":41,"tag":249,"props":596,"children":598},{"className":597},[],[599],{"type":47,"value":600},"fit exponential trendline to cumulative users; report growth rate b and doubling time",{"type":41,"tag":218,"props":602,"children":603},{},[604,608],{"type":41,"tag":245,"props":605,"children":606},{},[607],{"type":47,"value":484},{"type":41,"tag":245,"props":609,"children":610},{},[611],{"type":41,"tag":249,"props":612,"children":614},{"className":613},[],[615],{"type":47,"value":616},"overlay 3-month and 12-month moving averages on annual temperature data; highlight long-term trend",{"type":41,"tag":218,"props":618,"children":619},{},[620,625],{"type":41,"tag":245,"props":621,"children":622},{},[623],{"type":47,"value":624},"Multi-series",{"type":41,"tag":245,"props":626,"children":627},{},[628],{"type":41,"tag":249,"props":629,"children":631},{"className":630},[],[632],{"type":47,"value":633},"fit linear trendlines for each product in the 'product' column; overlay on one plot; table of slopes ranked highest to lowest",{"type":41,"tag":218,"props":635,"children":636},{},[637,642],{"type":41,"tag":245,"props":638,"children":639},{},[640],{"type":47,"value":641},"Log-linear",{"type":41,"tag":245,"props":643,"children":644},{},[645],{"type":41,"tag":249,"props":646,"children":648},{"className":647},[],[649],{"type":47,"value":650},"fit linear trendline to log(y) vs x (semi-log plot); report log growth rate; back-transform to original scale",{"type":41,"tag":42,"props":652,"children":654},{"id":653},"assumptions-to-check",[655],{"type":47,"value":656},"Assumptions to Check",{"type":41,"tag":658,"props":659,"children":660},"ul",{},[661,671,681,691,701],{"type":41,"tag":129,"props":662,"children":663},{},[664,669],{"type":41,"tag":56,"props":665,"children":666},{},[667],{"type":47,"value":668},"Trendline type matches the data",{"type":47,"value":670}," — exponential requires all y > 0; a linear fit to exponential data will show systematic curvature in the residuals",{"type":41,"tag":129,"props":672,"children":673},{},[674,679],{"type":41,"tag":56,"props":675,"children":676},{},[677],{"type":47,"value":678},"Extrapolation caution",{"type":47,"value":680}," — all trendlines can be extended beyond the data range, but reliability drops rapidly; polynomial fits in particular can diverge wildly outside the observed x range",{"type":41,"tag":129,"props":682,"children":683},{},[684,689],{"type":41,"tag":56,"props":685,"children":686},{},[687],{"type":47,"value":688},"Outliers inflate uncertainty",{"type":47,"value":690}," — a single anomalous point can drag the trendline significantly; ask the AI to fit with and without outliers to assess sensitivity",{"type":41,"tag":129,"props":692,"children":693},{},[694,699],{"type":41,"tag":56,"props":695,"children":696},{},[697],{"type":47,"value":698},"Time series autocorrelation",{"type":47,"value":700}," — standard R² and confidence intervals assume independent residuals; if observations are correlated over time (as they often are), standard errors are underestimated and confidence bands are too narrow",{"type":41,"tag":129,"props":702,"children":703},{},[704,709,711,717],{"type":41,"tag":56,"props":705,"children":706},{},[707],{"type":47,"value":708},"Seasonality is not a trendline",{"type":47,"value":710}," — if your data shows strong seasonal cycles, fit a trend + seasonality model (see the ",{"type":41,"tag":179,"props":712,"children":714},{"href":713},"/tools/time-series-decomposition",[715],{"type":47,"value":716},"Time Series Decomposition",{"type":47,"value":718}," tool) rather than a single trendline that will smear the cycles into the fit",{"type":41,"tag":42,"props":720,"children":722},{"id":721},"related-tools",[723],{"type":47,"value":724},"Related Tools",{"type":41,"tag":50,"props":726,"children":727},{},[728,730,736,738,742,744,750,752,758],{"type":47,"value":729},"Use the ",{"type":41,"tag":179,"props":731,"children":733},{"href":732},"/tools/polynomial-regression",[734],{"type":47,"value":735},"Polynomial Regression",{"type":47,"value":737}," tool for curved relationships where you want full statistical inference (F-tests, AIC-based degree selection, standardized coefficients). Use the ",{"type":41,"tag":179,"props":739,"children":740},{"href":713},[741],{"type":47,"value":716},{"type":47,"value":743}," tool when you need to separate the trend from seasonal cycles and residuals. Use the ",{"type":41,"tag":179,"props":745,"children":747},{"href":746},"/tools/linear-regression",[748],{"type":47,"value":749},"Linear Regression",{"type":47,"value":751}," tool when you have multiple predictor variables and want to control for confounders. Use the ",{"type":41,"tag":179,"props":753,"children":755},{"href":754},"/tools/logistic-curve-fit",[756],{"type":47,"value":757},"Logistic Growth Curve Fit",{"type":47,"value":759}," when your series is S-shaped and approaching a natural ceiling (carrying capacity).",{"type":41,"tag":42,"props":761,"children":763},{"id":762},"frequently-asked-questions",[764],{"type":47,"value":765},"Frequently Asked Questions",{"type":41,"tag":50,"props":767,"children":768},{},[769,774,776,781,783,788,790,794,796,801,803,808,810,814,816,821],{"type":41,"tag":56,"props":770,"children":771},{},[772],{"type":47,"value":773},"Which trendline type should I use?",{"type":47,"value":775},"\nStart with the underlying process: if your variable grows by a fixed ",{"type":41,"tag":78,"props":777,"children":778},{},[779],{"type":47,"value":780},"amount",{"type":47,"value":782}," per period (revenue increases by $5k/month regardless of the current level), use ",{"type":41,"tag":56,"props":784,"children":785},{},[786],{"type":47,"value":787},"linear",{"type":47,"value":789},". If it grows by a fixed ",{"type":41,"tag":78,"props":791,"children":792},{},[793],{"type":47,"value":82},{"type":47,"value":795}," (revenue increases 8% per month), use ",{"type":41,"tag":56,"props":797,"children":798},{},[799],{"type":47,"value":800},"exponential",{"type":47,"value":802},". If the relationship has a peak or valley (e.g. an environmental Kuznets curve), use ",{"type":41,"tag":56,"props":804,"children":805},{},[806],{"type":47,"value":807},"polynomial degree 2",{"type":47,"value":809},". If you just want to visualize the direction without a parametric commitment, use a ",{"type":41,"tag":56,"props":811,"children":812},{},[813],{"type":47,"value":96},{"type":47,"value":815},". Ask the AI to ",{"type":41,"tag":78,"props":817,"children":818},{},[819],{"type":47,"value":820},"\"fit all four types and show R² for each\"",{"type":47,"value":822}," when you're unsure.",{"type":41,"tag":50,"props":824,"children":825},{},[826,831,833,838,840,845],{"type":41,"tag":56,"props":827,"children":828},{},[829],{"type":47,"value":830},"What does a negative exponential growth rate mean?",{"type":47,"value":832},"\nIn the model y = ae^(bx), a negative ",{"type":41,"tag":56,"props":834,"children":835},{},[836],{"type":47,"value":837},"b",{"type":47,"value":839}," means exponential ",{"type":41,"tag":78,"props":841,"children":842},{},[843],{"type":47,"value":844},"decay",{"type":47,"value":846}," — the value decreases by a constant percentage per unit x. This is appropriate for radioactive decay, drug elimination, cooling, or any process where the rate of decrease is proportional to the current value. The half-life is ln(2)/|b| — the time for the value to drop to half its current level.",{"type":41,"tag":50,"props":848,"children":849},{},[850,855],{"type":41,"tag":56,"props":851,"children":852},{},[853],{"type":47,"value":854},"My R² is high but the forecast looks wrong — why?",{"type":47,"value":856},"\nHigh R² only measures fit within the observed data range. A polynomial that perfectly fits 10 historical data points can diverge wildly when extrapolated. This is especially common with degree 3+ polynomials. Always plot the projection region separately, and prefer exponential or linear trendlines for forecasting unless there is a strong reason for the curve to continue.",{"type":41,"tag":50,"props":858,"children":859},{},[860,865,867,873,875,881],{"type":41,"tag":56,"props":861,"children":862},{},[863],{"type":47,"value":864},"How do I get a trendline equation for Excel or Google Sheets?",{"type":47,"value":866},"\nAfter the AI fits the trendline, it reports the equation parameters (slope, intercept, growth rate, polynomial coefficients). You can enter these directly in a spreadsheet as a formula. For example, a linear trendline y = 5.7x + 3.1 becomes ",{"type":41,"tag":249,"props":868,"children":870},{"className":869},[],[871],{"type":47,"value":872},"=5.7*A2+3.1",{"type":47,"value":874}," in Excel. For exponential, y = 17.4×e^(0.078x) becomes ",{"type":41,"tag":249,"props":876,"children":878},{"className":877},[],[879],{"type":47,"value":880},"=17.4*EXP(0.078*A2)",{"type":47,"value":882},".",{"type":41,"tag":50,"props":884,"children":885},{},[886,891,893,898],{"type":41,"tag":56,"props":887,"children":888},{},[889],{"type":47,"value":890},"Can I fit a trendline to multiple groups at once?",{"type":47,"value":892},"\nYes — include a group or category column in your data and ask the AI to ",{"type":41,"tag":78,"props":894,"children":895},{},[896],{"type":47,"value":897},"\"fit a linear trendline for each group in the 'region' column; overlay on one plot; table of slopes and R² per group\"",{"type":47,"value":899},". This produces a panel of parallel trendlines useful for comparing growth rates across products, countries, or experimental conditions.",{"title":7,"searchDepth":901,"depth":901,"links":902},2,[903,904,905,906,907,908,909,910],{"id":44,"depth":901,"text":48},{"id":120,"depth":901,"text":123},{"id":205,"depth":901,"text":208},{"id":369,"depth":901,"text":372},{"id":524,"depth":901,"text":527},{"id":653,"depth":901,"text":656},{"id":721,"depth":901,"text":724},{"id":762,"depth":901,"text":765},"markdown","content:tools:049.trendline-calculator.md","content","tools/049.trendline-calculator.md","tools/049.trendline-calculator","md",{"loc":4},1775502471815]